Home | Biodata | Biography | Photo Gallery | Publications | Tributes [Back to Surveys List] Political - Socio - Economic-Surveys

Acknowledgement

The survey reported in part IV was conducted thanks to a grant from the University Grants Commission under the "Research Participation Programme for Post Graduate/Under graduate Students". The authors wish to thank Messrs K. Narayanan, S. Ranganathan, V. N. Govindan and K. Paul Devakumar, under-graduate students of Statistics who did the field work and Messrs N. Mahatvaraj and M. Kumaran, post-graduate students who processed the data. They are also grateful to Messrs Josef James and R. Rajkumar for their comments on an earlier draft of the paper. But the responsibility for the paper in its present form is only that of the authors.

I

The excessive enthusiasm to produce simple macro models of growth to identify the causes for the poverty of nations and to suggest quick remedies for the evils has led to rather distorted representations of basic functional relationships in the economy, especially in the so-called less developed countries (LDC's). One of the most glaring of these has been the assumptions usually made about the savings decision in the economy. Following macro models of the Keynesian type it is generally assumed that savings in the economy are a function of the levels of income, or per capita income. On this basis it was argued in the early days of the development debate that underdeveloped countries, defined as countries with low per capita incomes suffered from a chronic savings gap. In fact, this argument became the corner stone of the ' Vicious circle' theories of under-development at the academic level and the basis of the massive external assistance program for development at the policy level.

However, the need for disaggregated views of the economy both to understand its structural characteristics and to prescribe more effective policy measures has been stressed in more recent studies. In particular, it has become necessary to ask who are the savers and investors in the economy and why they behave the way they do. The main purpose of the present study is to examine some empirical evidence relating to these questions gathered from field studies conducted in some areas in South India. It is the belief of the authors that these questions are of major theoretical significance and that they also impinge upon strategies of growth. Hence, the paper also raises these theoretical issues, with the hope that they will pave the way for more, and more adequate empirical studies in the general area of savings and investment decisions. The main problems confronted in these areas today are not the lack of logical and sophisticated theoretical explanations but the paucity of empirical evidence to substantiate existing theories and to pose new ones.

II

It is generally assumed that economists today have all the information they require about the determinants of savings. From the rate of interest-related explanations of savings the profession has moved into the income-related explanations of savings. One may argue whether savings are related to absolute income, relative income or permanent income, but to Keynesian and post-Keynesian economists the main determinant of savings in the economy is the level of income. Or one may go back to the classical premises and insist that savings depend not on the level of income, but on its distribution among the social or economic classes, capitalists and workers, for instance. Most of the theories of savings today come within either of these two broad categories. There are, of course demographically based ' life-cycle ' theories of savings¹, but possibly because they explain the savings behaviour in terms of factors which are normally considered to be exogenous to economic analysis they do not have the kind of standing the income related theories of savings have.

Following the usual procedure we have classified income-related theories into the Keynesian and classical types. But parentage apart the distinction between these two consists in the extent of their aggregation: the Keynesian macro economic view presents a completely aggregated version of the economy², whereas there is an element of disaggregation in the classical theory. While in formal growth models the degree of disaggregation is an analytical property, the disaggregation in the classical theory came about, historically speaking, as a result of the classical writers'  sense of realism; it would have been impossible for Adam Smith, Ricordo, Mill and Marx to ignore the obvious class division in their society or to fail to take note of the distinctly differing propensities to save of the capitalists and the workers. Strangely enough, the absolutising of the classical principle that capitalists or entrepreneurs do not consume, and that workers do not save is a fairly recent analytical device made necessary for the convenience of formal growth models.³ In this sense the ' classical ' theory of saving is post-Keynesian, and must be seen within the general context of the Keynesian, income determination system, or as its natural extension.⁴

A legitimate question to raise here is whether in terms of the realities of a given situation (as it exists in a country like India for instance) or in the terms of analysis, disaggregation into two sectors is sufficient to deal with the problems of savings. To answer this question it is necessary to trace the nature and purpose of the Keynesian aggregation and the classical disaggregation. It is clear that in Keynes's initial formulation of the consumption (savings) function the main objective is not the study of consumption (saving) at all. Opening his discussion of the new concept of ' propensity to consume ' Keynes makes it very explicit that ' the ultimate object of our analysis is to discover what determines the ' volume of employment.' ⁵ His attempt, therefore, is not to discover the determinants of consumption, but to discover how ' the sum which will be spent on consumption when employment Is at a given level, and the sum which will be devoted to investment' ⁶ together constituted the ' Proceeds ' which that level of employment is expected to realize. In other words, Keynes's primary concern is with the aggregate demand function, not the consumption function. If this simple fact were appreciated many of the subsequent controversies about the nature of the Keynesian consumption function could have been avoided. As for consumption itself, Keynes possibly has a longer list of objective and subjective determinants (which are dealt with in chapters 8 and 9 of the General Theory) than are brought into the discussion these days. And if this statement that ' the amount that the community spends on consumption obviously depend (i) partly on the amount of its income, (ii) partly on the other objective attendant circumstances, and (iii) partly on the subjective needs and the psychological propensities and habits of the individuals composing it and the principles on which the income is divided between them (which may suffer modification as output is increased)⁷ is taken as his formulation of the theory of consumption, it is far more catholic than is generally conceded. But this is not the point. What is important is to recognise that for Keynes the main problem was to establish that the sum spent on consumption is a function, and a fairly stable function of income. In his words again ' Windfall changes in capital-values will be capable of changing the propensity to consume, and substantial changes in the rate of interest and in fiscal policy may make some difference but the other objective factors which might affect it, while they must not be overlooked, are not likely to be important in ordinary circumstances. In fact that, given the general economic situation the expenditure on consumption in terms of wage units depends in the main, on the volume of output and employment is the justification for summing up the other factors in the portmanteau function ' propensity to consume '. For whilst other factors are capable of varying (and this must not be forgotten), the aggregate income measured in terms of the wage unit is, as a rule, the principal variable upon which the consumption constituent of the aggregate demand function will depend.⁸

All this and much more is simply to prove that the level of income is determined not by the sum spent on consumption, but by the sum spent on investment. And to emphasize further the importance of the investment decision Keynes has also to assume that the investment decision and the consumption (savings) decision are independent. This was partly a generalization of what was an empirical reality in his case the savings decision was made by the households and the investment decision by the entrepreneurs. What was implicit in Keynes was made explicit by Beveridge, according to whom the Keynesian system depends, among other things, on the assumption that (a) decisions to save and decisions to invest are made by different sets of people at different times, and for different reasons and may thus get out of step, and (b) the amount which any community may try to save is not governed by the opportunity of investment, but by the total income of a community and its distributions.⁹

In the ' classical ' models, of course, savings and investment are both made by the same decision-makers, capitalists. But even in these models the independence of the savings (consumption) and investment decisions is maintained by making the savings decision a function of the decision of income between capitalists and workers and the investment decision to depend on other factors such as the capitalists' desire to accumulate. In the extreme case of the classical model where workers only consume and capitalists only invest, the independence of the consumption (savings) decision and the investment decision is absolutized.

Because of the convergence of the Keynesian and classical models on the question of the independence of these two decisions, this property has assumed a certain generality and finality in theoretical discussions. Exceptions, if any, have come largely in the context of empirical studies of savings and investment. At least three such studies have claimed attention in recent years: the Friend and Kravis study of entrepreneurial income, saving and investment in the United States, Houthakkar's international comparison of personal savings and Kelley and Williamson's Indonesian case study of household savings behaviour.¹⁰ One of the concerns implicit in all three studies is expressed in the words of Friend and Kravis: ' The study shows the potential danger in constructing aggregate saving or consumption functions ... and in assuming that savings and investment decisions are independent for individuals as a whole.' ¹¹ This finding is prompted by the evidence emerging from all these studies that the savings pattern of entrepreneurial households is distinctly different from that of others. In all the studies savings from entrepreneurial incomes is seen to be considerably higher than the average savings-income ratio of households. While Houthakker is concerned only with the relationship between savings and the sources of income of income recipients, the other two studies point out also that the savings-income ratio of the self-employed households is higher than that of other households  They also show that the entrepreneurial households have a higher marginal propensity to save. Using simple linear functions of the form S=a+bY (in the Friend and Kravis study) and of the form S/N = a+bY/N (in the Kelley and Williamson study) where S = Savings, Y = Income and N = family size they show that the b coefficient is higher in the case of the self-employed than for other kinds of households. In one comparison in the Friend and Kravis study the b coefficient is 0.443 for the self-employed compared to 0.216 for all others, and in another, unincorporated business owners have a b coefficient of 0.545 as against 0.246 for all others. The estimates of the coefficient b in the Kelley and Williamson study are as follows: Farmer, 0.167, Trader and Craftsman, 0.426, Owner of business, O.308, Government Employee 0.048, other Wage earner, 0.111. While these differences are themselves impressive, the more important point is to note that they arise from the differences in the conditions of and motivations for savings of the entrepreneurial households. Noting a close connection between entrepreneurs' savings and their net investment Friend and Kravis infer that ' (a) entrepreneurial savings is motivated by the need or desire to finance investment and (b) business investment is a ready-made offset to a kind of saving that is an important fraction of total personal savings.' ¹² After noticing that the propensity to save out of entrepreneurial income is significantly higher than the propensity to save out of wage income, Houthakker also thinks that it may be the result of the entrepreneurs having direct access to investment opportunities. Kelley arid Williamson offer the following explanation for the difference between the behaviour of wage-earning households and self-employed households: ' For the wage-earning households, which offers only its labour service to the factor market, the determination of savings involves, in addition to an allocation between present and future consumption, a decision regarding the maintenance of the existing stock of human capital and the increments in that stock. The self-employed entrepreneurs, on the other hand, receive income for labour services, for the use of non-human earning assets, and for managerial abilities, to the extent that household savings decision are simultaneously determined with those based on 'entrepreneurial earning assets, their different consumption behavior compared to other occupational households is to be expected.' ¹³

Two related conclusions follow from these studies: the savings decisions of producer-households may depend largely on their investment opportunities, and hence for a group of decisions makers in the economy the savings (consumption) and investment decisions are not independent. The phenomenon dealt with in these studies, therefore, lies outside the conventional Keynesian and classical models of savings. The significance of these findings, of course, will depend on the magnitude of entrepreneurial savings and its proportion to total savings. Friend and Kravis consider their findings interesting enough to suggest further investigations enough to suggest further investigations. Kelley and Williamson dealing with the Indonesian case suggest that the phenomenon is likely to be of special significance in the LCD's ' where the entrepreneur assumes a far greater role as a result of the relative size of the agricultural sector and also of the relative backwardness of the corporate movement in the nonagricultural sector.' ¹⁴ However they rightly regret the dearth of empirical evidence on household savings in such economies.

III

This question has received some attention although far from adequate in India. Like many other LDC's, especially in the South Asian region,¹⁵ the share of the household sector in gross domestic savings in India has been very high. According to official statistics between 1950-51 and 1962-63, the households' share in domestic savings ranged between 52.7 per cent in 1951-52 and 82.3 per cent in 1955-56 with a mean of 74.4.¹⁶ But at the official level hardly any attempt has been made to go beyond the usual breakdown of total savings into the Government, Corporate and Household Sectors except that the household sector has been further divided into rural and urban. Even some of the leading non-official estimates of savings in the economy like that of the National Council of Applied Economic Research have been undertaken within such conventional classifications.¹⁷ In some of the other studies, however, the NCAER has paid special attention to the source of income and occupation as determinants of savings. In its Urban Income and Savings based on a country-wide sample survey conducted in 1959-60 it is recorded that ' whether a person works for himself (self-employed) or for another seems to lead to significant differences in the savings behavior.'  The self-employed as a class according to the present study save 1 1/2 to 2 times as much a proportion of their income as employees.' ¹⁸ Another study by NCAER found that 50 per cent of the households surveyed expressed a preference for investing their savings in physical assets, and that households directly engaged in business have a strong preference for further investment in their business.¹⁹ But neither study attempted a more detailed analysis of these findings.

The theoretical and policy implications of the high propensity to save of the producer households and their equally strong propensity to invest the savings directly in their own production concerns, however, have been examined by a few individual scholars. In one of the early attempts to produce an estimate of savings in the economy Mazumdar noted the dominating role of unincorporated  enterprises in total savings.²⁰ In the late fifties and early sixties Malenbaum made a systematic study of consumption and savings in India using mainly the data by the National Sample Survey and also information collected directly through field studies.²¹ According to him ' a nation of entrepreneurs and most of India's work force is self-employed has a high propensity to invest directly any surpluses arising in the individual enterprise. Most Indian enterprises tend to have this direct investment attitude.' ²² His main line of argument was to point out that ' the high propensity to invest directly ' of the largest group of savers in the economy set fairly well-defined limits to the mobilization of savings in the economy, and consequently on the size of the public sector, but he did not examine whether the propensity to invest directly had any influence on the savings decisions of the kind of savers he was studying. On the question of the determinants of savings in the economy he preferred to depend on macro-economic explanations such as the output of food grains.²³

A more recent analysis of the savings behavior of producer households and its implications for public policy has been Kurien's study.²⁴ Kurien's approach to this issue has  been a follow up of his earlier studies of the structure of the Indian economy where he has tried to show that in a labour- abundant economy such as India the presence of a group of producers who would use their labour and non-labour resources together is inevitable.²⁵ Such producers' propensity to invest directly he envisages as the dynamic element of the kind of model he has postulated for the Indian economy. With the help of a new typology of decision-makers in the economy, he has pointed out that an ' Indian system' dominated by 'own-investors, savings-users and self-employers' is analytically distinct from the Keynesian and Classical systems, neither of which has such decision-makers in them.²⁶ He argued further that the determinant of saving in the Indian system could neither be the level of income (as in the Keynesian system) nor the distribution of income (as in the classical system), and that unlike in these two systems, the consumption (savings) and investment decisions in the Indian System could not be treated as independent. In fact, he featured the inter-dependence of the consumption (savings) and investment decisions as one of the distinguishing features of the India system. He expressed the relationship between income, consumption, savings and investment in such a system as
C= Y [S = f (IO) ],
where. Y, C and S stand as usual for income, consumption and savings, IO the investment opportunity and a minimum level of consumption determined by the decision-makers themselves.²⁷ He, therefore, hypothesized that the level of savings, given income, was a positive function of investment opportunity, implying a built-in 'consumption squeeze', mechanism in the Indian System.

For empirically testing the hypothesis, two inferences were drawn from it: (i) the main motivation for savings by producer households is their desire to expand their own business concerns, and (ii) when producer households take on expansion schemes or have to repay business loans, they effect substantial reductions in their domestic consumption expenditures. A sample survey of producer households in. Madras City was conducted early in 1967 and its findings are reported in Indian Economic Crisis, Chapter V.

To make a quantitative study of the problem the survey attempted to gather precise information about the initial and current capital stock of the respondents, their incomes and expenditures, nature of expansion work etc. It could not be claimed that the quantitative aspect of the survey was altogether satisfactory. But the study showed on the whole a very high savings-income ratio for the producer households, with wide variations depending on the levels of income. A regression of savings as a function of income gave R² = 0.82 with a marginal propensity to save of  0.75. One of the main intentions of the survey was to compare the savings-income ratio of those who were conducting expansion work and those who were not. But as it turned out, at the time of the survey hardly anybody had an ongoing expansion programme and almost all producers were producing below capacity. Although this appeared strange at the time of the survey, an adequate explanation came soon enough in the form of the industrial recession of 1967-68.

Direct questions addressed to the respondents seemed to support the hypothesis. It was seen that 90 per cent of the households gave first ranking to expansion of business as the purpose of business for savings and that 77 per cent of households depended solely or mainly on their own savings for business expansion. To the question: ' Do you save more than usual when you are expanding your business or when you have to repay business loans?' affirmative answer came from 65 per cent of the households.

IV

Another study was undertaken to test the hypothesis further. It consisted of an extensive survey of the Ramnad district of Tamil Nadu, the erstwhile Madras State, conducted during the summer months of 1968. Ramnad district is mainly dry due to the lack of rain and the absence of adequate irrigation facilities. But this fact has perhaps, stimulated the growth of a number of ' small industries' in the region.²⁸ A full list of small Industrial units in the district numbering 462 is available in the Directory of Small Scale Industrial Units in Madras State, 1965 , a publication of the State Government's Department of Industry. The Directory, however, does not make any reference to the ownership pattern of these units ranging from individual proprietorships to incorporated companies. The attempt of the survey was to trace the " household production units" among them individual ownerships and partnerships of family members and to study their savings-investment relationship, concentrating on the motivations for savings. Over 90 per cent of the small industrial units in the District was located in the nine small towns of Sivakasi, Sattur, Virudunagar, Tirupattur, Rajapalayam, Karaikudi, Aruppukottai, Paramakudi and Srivilliputtur. All units listed in these towns were contacted. Some units had closed down since the publication of the Directory; some were registered companies with limited liability. In all about 270 units came within the category of ' household production units' as defined above. Not all these units, however, cooperated fully in answering the questionnaire. Most of these units, especially in Sivakasi were producing match boxes and fire-works; Virudunagar had a concentration of coffee powder manufacture; but there were also units producing groundnut oil, camphor, tin containers, tin lamps, alloy metal powder, household vessels, biscuits and pipes.

While organizationally these units were classed as household production units, it is important to note that functionally they were primarily business units, not merely households doing some side-business. In the case of 64 per cent of these households, the kind of business they were doing was the only source of their income, while the rest had business income as the main source supplemented by income from land and agriculture. Over 88 per cent of them conducted business in place other than their houses arid almost all of them (98.9 per cent) used some hired labour. Asked whether they had any kind of accepted financial year for their operations and accounts, 97.1 per cent mentioned the April-March year which is the financial year for the government and for most large scale business concerns. The initial capital of these units varied considerably, the lowest being Rs.660 and the highest Rs. 1, 200, 000, with a little over 40 per cent having an initial investment of less than Rs. 10, 000, These units, therefore, were small business concerns which have come up because of the initiative and enterprise of some individuals and families in a "nation of entrepreneurs". One other aspect which deserves attention regarding the organizational aspect of these units is the fact that most of them ( 96.6 per cent) were independent units in the sense that they had no organizational connection with any other production units. The rest had some form of subcontract arrangements with other concerns.

The respondents were asked to list the reasons for their savings. The open ended question suggested the following reasons: (i) Expansion of business, (ii) Purchase of property, (iii) Education of children, (iv) Repayment of loans, (v) Buying National Certificates. Of the 244 who responded, 230 mentioned expansion of business as a reason for saving; 123 included purchase of property also; repayment of loans figured in 25 responses, education in 17 and buying of National Savings Certificate only in one. The dominating role of business considerations in savings was correlated with reference to the utilization of past savings. The following table gives the details.

TABLE I:  RANKING OF UTILIZATION OF PAST SAVINGS

	Ranking
	I	II	III	IV
Invested in business	214			1
Purchase of Property		37	9	1
Insurance		63	12	1
Savings Account in Banks		8	6	7
Loaned to others		2	5	1
Fixed Deposit		1	1
Cash			3	3
Securities			3

Only 94 units of the total sample undertook any expansion programme in the two years immediately preceding the survey. The patterns of financing of these expansion schemes appeared to be somewhat different from the one suggested above. Only 19 out of the 94 depended on own savings for the recent expansions. Credit from cooperatives was the source of financing for the largest number, 48. The commercial banks helped 13 units and the money-lender 7. The other expansion programs were financed by loans from relatives or friends and by the sale of property. It is not clear whether this was an indication of credit from outside becoming the main source of business expansions. Considering the fact that the majority of the producers did not have any expansion projects in the immediate past, it may simply mean that cooperative credit enabled a large number of units to grow which may not have undertaken any expansion but for such help. It may be pointed out too that among the 15 units which had expansion programmes at the time of the survey (221 units did not have any expansion programme) 8 depended on their own resources, 3 on cooperative credit, 1 on bank loan and the rest on other forms of borrowing including 1 from the money lender. Although there is no quantitative precision about these findings they lend support to the hypothesis that the producer households depend largely on their own resources for their business and its growth and that the financing of investment in own business is the main reason for their savings activity.

The second question to be examined was whether the need and opportunity for investment had any impact on the consumption (saving) decision of these units. There were 240 affirmative answers and 35 negative ones to the question, "Do you save more than usual when you are expanding your business or when you have to repay business loans". There was another question also which was meant to throw light on this matter. The respondents were asked; "If you are given a partial subsidy to expand your business units how will you meet the rest?" The question suggested the following possibilities: postponing major purchases, reducing expenditure, selling things like jewellery and property, getting loans from others, and working longer hours. The respondents were free to indicate as many possibilities as they wanted. The following tabulation is quite revealing:

Getting loans from others	192
Selling jewellery, property etc	151
Reducing Expenditure	94
Working longer hours	23
Postponing major purchases	22
Taking another partner	16
Other	11

Reducing of expenditure gets only the third ranking in the tabulation with loans having a pronounced lead. But it must be remembered that reducing expenditure was given as a prominent possibility even when "easier" options were available in the question and that both working longer hours and postponing major purchases are, along with the direct reduction of expenditure activities closely related to the savings decision. Even where resorting to borrowing is the first response to an investment opportunity, as seen already, the households have the habit of reducing normal consumption expenditure to repay the business loan. Hence making adjustments in the normal pattern of consumption must be seen as a usual pattern of response to opportunities to expand business activity.

To further verify this possibility a more detailed analysis of the consumption pattern of the units was attempted. No effort was made to get precise quantitative estimates of monthly consumption expenditures or their break down. Instead, the units were asked to rank the major items of their consumption. The expenditure pattern is indicated in Table 2.

TABLE 2:  EXPENDITURE PATTERN

Items of domestic expenditure	Rank					Total
	I	II	III	IV	V
Food	269					269
Clothing		176	92			268
Education		90	150	6	13	259
Medical Bill		3	8	163	85	259
Entertainment			16	93	154	253
Total	269	269	266	262	151

The unite were also asked to state whether they would be able to reduce their normal expenditures in times of need and to indicate the extent to which the reduction could be made. They were further asked to identify the particular items on which the reduction could be effected. Tables 3 and 4 bring together the answers to these questions.

TABLE 3   POSSIBLE REDUCTION IN DOMESTIC EXPENDITURE

Percentage of Expenditure	Frequency Numbers	Percentage
0	46	018.9
1	1	000.4
4	1	000.4
5	39	016.0
6	3	001.2
7	3	001.2
10	70	028.8
12	4	001.7
15	29	011.9
20	27	011.1
25	10	004.1
30	10	004.1
Total	243	100.0

TABLE 4  ITEMS OF DOMESTIC EXPENDITURE WHERE REDUCTION IS POSSIBLE

Items	Frequency
Food	11
Clothing	79
Medical	3
Education	1
Entertainment	171
Tour expenses	1

Table 3 gives an idea of the ' built-in consumption squeeze mechanism' of the producer-households. It is impressive that more than 80 per cent of the households indicated that they could reduce their expenditures by 5 per cent or more and that over 60 per cent of them could bring about a more than 10 per cent reduction in their usual consumption. This is an indication of the tremendous elasticity of savings among the group of people who are the habitual savers in the economy. And Table IV shows also that the households have a well thought out scheme of priorities when it comes to reduction of their normal expenditures for the sake of the growth of their business concerns.

V

Even the evidence marshalled here is not adequate to give conclusive verification of the built-in consumption squeeze hypothesis. The empirical verification of the hypothesis is in fact a difficult task because in a large and diverse country like India it is also necessary to find out whether the features observed in this study are only special regional characteristics. But the evidence coming from the American study of Friend and Kravis, the international study of Houthakker, the Indonesian study of Kelley and Williamson, the studies in South India reported here, and one or two other field studies that we know of ²⁹, seem to suggest that investment in their own business is the chief motivation for savings of the producer households and that their consumption (savings) decisions are closely related to investment opportunities and needs.

If this is right, there are certain implications both for economic policy and for economic theory which can only be briefly spelled out here.³⁰ One of the most obvious of these pertains to the procedures of collection and presentation of official statistics on savings and investment in countries like India where economic activity is dominated by producer households. The conventional procedure of estimating domestic savings dividing the economy into the Government Sector, the Corporate Sector and the Household Sector presupposes that the motivations for savings in these sectors are different. The households are clubbed together on the (implicit) assumption that households save, but do not invest. But in countries like India where the ' Household Sector ' is the residual sector consisting of all savers other than the Government and the corporations and where the vast majority of households are producer households, there is at least a strong case for dividing the household sector itself into two - producer households and non-producer households and making the former the residual category. Such a reclassification will considerably enhance our understanding of the patterns of savings in such economies.

In turn the new understanding may raise many questions about the problem of growth of economies where, producer households are a significant economic category. It will certainly challenge theories which feature the inelasticity of savings as the only or the main cause of underdevelopment in many parts of the world. It may also make it necessary to review savings-oriented growth strategies. The Indian Five Year Plans, for instance, have been basically investment allocation plans on the assumption that a great deal of the savings in the economy (mainly from the household sector) could be converted into a central pool of savings through appropriate measures of resource mobilization, taxation, inflationary financing and the like. If the patterns of savings in the economy are different from what these policy measures presuppose, such policies themselves may be largely responsible for the retarded growth experienced during the past two decades.

At the theoretical level two questions are worth considering. The first is the need to drop the savings function in growth models. There are some hints of this in Schupeter's  Business Cycles, a possibility also referred to in more recent writings,³¹ but still not fully explored mainly because of the difficulties of formalizing investment functions. The profession also does not seem to have enough empirical observations about investment functions, especially of the type of investors discussed in the present study. Both these are pressing areas for further research.

The existence and activities of the producer households raise also more fundamental theoretical questions. The producer households make all conceivable economic decisions-production, consumption, savings, investment. The main body of economic theory as we have it today has as its basic postulate decision-makers who make each of these decisions independently of other decisions. The assumption of independent decisions-making underlies both micro and macro theories. The producer households are in this sense not only decision-makers; they are virtually sub-economies. To understand their role and the nature of economies themselves is it not necessary to have a new conceptualization of the 'economy' not as consisting of functionally differentiated independent decision-makers but as consisting of a cluster of multi level decision-makers?

REFERENCES

1. E. Modigliani and A. Ando, " Tests of the Life-Cycle Hypotheses of Savings', Bulletin of the Oxford. Institute of Statistics, (May, 1957); and Modigliani and Ando, ' The Life Cycle Hypothesis of Savings', American Economic Review, (Vol. LIII, No. I, March, 1963)
2. While Keynes related the aggregate level of savings (or more accurately consumption) in the economy to the level of income, in the writings of Duesenberry and Friedman which have given rise to the relative income and permanent income hypothesis the discussion is about the savings propensity of individuals. It may be pointed out that at the level of individuals Keynes recognised a number of  'subjective' factors as being responsible for the savings (or rather consumption) decisions. See General Theory, Ch.9
3. The 'classical savings function' in its rudimentary form can be traced back at least to Ricardo, but it has come into prominence only recently in the so-called 'Cambridge models of growth', especially in the writings of Kahn, Kaldor and Robinson.
4. This point is particularly stressed by Kaldor in his 'Alternative Theories of Distribution', Review of Economic Studies (Vol. XXIII, 1956)
5. General Theory, p. 89, emphasis added
6. Ibid, p. 89
7. Ibid, pp.90-91
8. Ibid, pp.95-96
9. Full Employment in a Free Society, London, George Allen and Unwin, Second Edition, 1960, p. 94
10. Irwin Friend, Irving Kravis, 'Entrepreneurial Income, Saving and Investment', American Economic Review  (Vol. XLVII, No. 3 June, 1957); H.S.Houthakker , 'An International Comparison of Personal Savings', Bulletin of International Statistical Institute, 38, and 'On some Determinants of Savings In Developed and Underdeveloped countries', in E.A.G. Robinson ed., Problems in Economic Development, London, Macmillan and Co., 1965; Allen C. Kelley and Jeffrey G. Williamson,  'Household Saving Behavior in Developing Economies: The Indonesian Case', Economic Development and Cultural Change, (Vol. 16, No, 3, April 1968).
11. Loc. cit., p. 290 .
12. Loc. cit., p. 289.
13. Loc. cit., pp. 388-389
14. Loc. cit., p. 389.
15. According to an ECAFE publication the households' share in gross domestic savings in 1959 in this region ranged between a low of 57 per cent for the Philippines and a high of 114 percent for South Korea, (United Nations, ECAFE, Review of Long Term Projections for Selected Countries in the ECAFE Region. Development Programming Techniques, Series No. 5, Bangkok, 1964).
16. D.R. Katkhate and. D.L. Deshponde, 'Estimates of Savings and Investment in the Indian economy: 1950-51 to 1962-63', The Reserve Bank of India Bulletin, March 1965.
17. For instance see, National Council of Applied Economic Research , Savings in India, 1965.
18. NCAER, Urban Income and Saving, p. 82
19. NCAER, Attitude Towards and Motivations for Savings, p. 23
20. H. Mazumdar, Business Savings in India, Bombay, Vora and Co., 1959.
21. Wilfred Malenbaum, Prospects for Indian Development, London, George Allen arid Unwin, 1962 and papers mentioned therein.
22.W. Malenbaum, 'How Large the Public Sector, 1961-66', Economic Weekly, Special Number, June 1959.
23. W. Malenbaum, op. cit., p. 245.
24. C.T. Kurien, Indian Economic Crisis : A Diagnostic Study, Bombay, Asia Publishing House, 1969
25. C.T. Kurien, A Theoretical Approach to the Indian Economy, Father P. Carty Endowment Lectures of the Madras University, 1966 (Bombay, Asia Publishing House, 1970).
26. Kurien, Indian Economic Crisis, Ch. IV
27. Ibid., p. 37
28. This seems to support the main hypothesis put forward by T.S. Epstein in Economic Deve1opment and Social Change in South India, Manchester, Manchester University Press, 1962. One of Ramnad's neighbouring districts is Thanjavur ( Tanjore) 'the granary of the South' and the home of the green revolution in paddy.
29. Cf. P.G.K. Panikar, Rural Savings in India, Bombay, Somaiya Publications, 1970; also the unpublished research work of R. Rajkumar, (who was the principal investigator of the Madras Study reported in Part III, and who has subsequently conducted a more detailed study of the savings behavior of households in selected parts of South India).
30. See especially, F.H, Hahn and R.C.O. Matthews, ' The Theory of Economic Growth: A Survey', Economic Journal, (Vol. 74, No.4, Dec.1964), particularly Sec.L.5.