Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital....

49
Trends in Rural Savings and Private Capital Formation in India SWP382 World Bank Staff Working Paper No. 382 April 1980 Prepared by: Raj Krishna and G.S. Raychaudhuri, Consultants South Asia Programs Department Copyright © 1980 The World Bank 1R, H Street, N.W. iington, D.C. 20433, U.S.A. views and interpretations in this document are those of t 1 i7 D irs and should not be attributed to the World Bank, to ii JI=J fl lID) \\/7 r ,ed organizations, or to any individual acting in their beh U UL D (( ,t2J 1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized

Transcript of Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital....

Page 1: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

Trends in Rural Savings and Private CapitalFormation in India

SWP382

World Bank Staff Working Paper No. 382

April 1980

Prepared by: Raj Krishna and G.S. Raychaudhuri, ConsultantsSouth Asia Programs Department

Copyright © 1980The World Bank

1R, H Street, N.W.iington, D.C. 20433, U.S.A.

views and interpretations in this document are those of t1 i7 D

irs and should not be attributed to the World Bank, to ii JI=J fl lID) \\/7 r,ed organizations, or to any individual acting in their beh U UL D (( ,t2J 1

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Page 2: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

The views and interpretations in this document are those of the authors and

should not be attributed to the World Bank, to its affiliated organizations,

or to any individual acting in their behalf.

WORLD BANK

Staff Working Paper No. 382

April 1980

TRENDS IN RURAL SAVINGS AND PRIVATE CAPITAL FORMATION IN INDIA

This paper reports on trends in rural household savings, net private

capital formation in agriculture and the growth and composition of tangible

net wealth in rural India from 1950-1973. It also presents fresh estimates of

the marginal and average propensities to save and invest during the period

with respect to measured as well as permanent and transient income. The

results show that there has been an acceleration of saving and capital form-

ation in more recent years.

Prepared by: Raj Krishna and G. S. Raychaudhuri, Consultants.South Asia Programs Department

Copyright (c) 1980The World Bank1818 H Street, N.W.Washington, D.C. 20433, U.S.A.

Page 3: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

TRENDS IN RURAL SAVINGS AND PRIVATE CAPITAL FORMATION IN INDIA

TABLE OF CONTENTS

Page No.

1. Introduction ................................................. 1

2. Growth of Saving and Capital Formation ....................... 3

3. Marginal Propensities to Save and Invest ..................... 9

4. Propensities in the Fifties and Sixties ...................... 14

5. Saving Out of Permanent and Transient Income .... ............. 19

6. Some Disaggregative Estimates ................................ 23

7. Tangible Wealth and Asset Composition ........................ 26

Appendix A: The Rural Household Savings Series .... .............. 34

Appendix B: Net Capital Formation in Agriculture Series ......... 38

Appendix C: Tangible Wealth of Rural Households .... ............. 39

Bibliography ..................................................... 41

Page 4: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

TRENDS IN RURAL SAVINGS AND PRIVATE

CAPITAL FORMATION IN INDIA1950-51 - 1973-74

by

Raj Krishna @G. S. Raychaudhuri @

1. Introduction**

This paper reports on trends in (a) rural household savings,

(b) net private capital formation in agriculture and (c) the growth and

composition of tangible net wealth in rural India during the 24-year period

1950-51 to 1973-74. It also presents fresh estimates of marginal and average

propensities to save and invest during this period with respect to measured

as well as permanent and transient income. Previous time series studies have

covered the earlier period 1950-51/1962-63. (Gupta, 1970, and Datta Roy

Choudhury, 1968.) The saving propensities estimated in one of these studies

are comparable with our estimates for this earlier period. But there has been

an acceleration of saving and capital formation in more recent years, and,

therefore, the saving propensities estimated by us for the recent period

1960-61/1973-74 are distinctly higher.

* The first draft of this paper was prepared for the World Bank.

** Thanks are due to Chandra Mazumdar and Anil K. Gupta for competent

statistical assistance, and to Sartaj Alam Kidwai for computer

programming.

The authors are grateful to an anonymous reviewer for comments which

induced the addition of Section 6 to an earlier version.

@ Professor of Economics and Reader in Economics respectively in Delhi

School of Economics.

Page 5: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 2 -

The paper is organized as follows.

The next Section presents the basic data on the growth of saving

and private capital formation - aggregate and per capita - in different Plan

periods, and particularly in the period since the start of the Green Revolution

in 1967. Variations in the average saving rate and the average rate of private

capital formation are also analyzed.

Section 3 contains estimates of the marginal propensities to save

(MPS) and invest (MPI) 1/.

From the computed functions we have estimated the positive intercept

on the income axis which measures the income level at which positive saving

begins. We call it the "critical minimum income" necessary for income to ex-

ceed consumption. The relationship of this critical income to the poverty

line is very instructive.

The fourth Section compares our time series estimates of saving pro-

pensities with those of earlier time series studied. Some all-India cross-

section saving rates estimated on the basis of data collected by the National

Council of Applied Economic Research have also been reviewed.

The fifth Section separates savings out of permanent and transient

income. The MPS out of permanent income appears to be higher than out of

measured income. Transient saving seems to augment or reduce permanent saving

by 8 to 9 percent on the average.

Section 6 presents some evidence on variations in saving rates across

different regions and the returns to investment in modern inputs and assets.

1/ We shall hereafter use "investment" as an abbreviation for privatecapital formation in agriculture.

Page 6: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

Section 7 analyzes the growth of reproducible tangible net wealth

(RTW) in the rural areas on the basis of comparable estimates for 1950, 1961,

1966 and 1972 developed by the Central Statistical Organization (CSO) and the

Reserve Bank of India (RBI). Gini coefficients, available for 1961-62 and

1971-72, roughly measuring the degree of inequality in the distribution of

RTW, are also reported. They indicate no significant change in the degree

of inequality.

Livestock Census data contain considerable information on the asset-

composition of phiysical capital. Trends in this composition are also dis-

cussed in Section 6.

2. Growth of Saving and Capital Formation

Two basic time series - net rural household saving (S), and net

domestic private capital formation in agriculture (NDCFA) - for the period

1950-51 to 1973-74 have been developed in Table 13 and plotted in Figure 1.

The saving series includes RBI figures for rural household saving up to 1959-

60, and CSO figures for the remaining 14 years 1/. The NDCFA series is

obtained from the National Income/Accounts Statistics of the CSO 2/. Private

capital formation includes construction, machinery and equipment, livestock

and change in stocks.

1/ The latter have been computed so as to be comparable with the earlier

RBI figures. (See Appendix A.)

2/ Some adjustments have been made. (See Appendix B.)

Page 7: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

Rural Saving, Private Investment and Income(Per capita in 1960-61 prices)

Rs. I ndia, 1950-51/1973-7435

yr - Per capita rural income (Rs).ndcfa - Per capita private invastment in agriculture (Rs).

s - Per capita rural saving (Rs).

30 y

- -- - /~~~001- 00

25

20

1 5 ndcfa -

1 0 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974

Financial year beginning April

World Bank-20516

Page 8: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

The average rural saving rate series is shown in Table 1 and plotted

in Figure 2 (a). It will be seen that the saving ratio has recorded a dis-

appointingly modest upward trend. Over the whole 21 year period it averaged

only 2.8 percent. In the first two Five Year Plans it averaged 2.3 percent.

It rose to 2.5 percent in the Third Plan and then again to 3.8 percent in the

Fourth Plan period. A significant step-up in the ratio occurred in the years

beginning 1965-66 when price support and the new technology were introduced.

The ratio increased from 2.5 percent in the Third Plan to 3.5 percent in the

eight years beginning 1966-67. But even after this increase, the level of the

saving rate remains very low indeed. The rural sector is still apparently

unable to save even 5 percent of its income - the ratio which Arthus Lewis

(1954) characterized as the typical indicator of a backward pre-development

economy.

For an explanation of this low level we must recall the fact that

at least 41.4 percent of the rural population remained below the poverty line

(Rs 42.45 per capita per month) even in 1973-74 on the basis of consumption

(a) The income series used for calculating the saving ratio is the sum of

the long-period rural savings series constructed by us and the ruralcomponent of the CSO estimate of aggregate consumption. Aggregate CSO

consumption is broken up into rural and urban consumption on the basis

of the rural: urban consumption ratio implicit in the National Sample

Survey data for each year.

Although the NSS consumption series are considered more satisfactory,

experiments with an alternative rural income concept in which this

series is used yielded inconsistent and incredible results for the two

periods 1950-51/1962-63 and 1960-61/1973-74. This outcome is clearly

due to the fact that the CSO and RBI saving estimates are related to

CSO income estimates and not to NSS data. Therefore, the choice was

made to use CSO consumption data to get the rural income series. This

choice has produced consistent and meaningful results for the whole

period 1950-51/1973-74. These results are also in harmony with other

bodies of independent cross-section data on rural saving, rural capital

formation etc.

Page 9: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 6 -

TABLE 1

RATIO OF RURAL SAVING AND INVESTNENT TO INCOME: INDIAPER CAPITA (1960-61 PRICES)

AverageNet Saving/ of Plan Net Investment/ Average of

Year Rural Income Periods Rural Income Plan Periods

(Percent)

1950-51 2.35 2.35 1.07 1.07

1951-52 2.40) 1.69)

1952-53 2.24) 0.91)

1953-54 2.36) 2.27 1.56) 1.58

1954-55 2.22) 1.67)

1955-56 2.11) 2.06)

1956-57 2.42) 3.30)

1957-58 2.03) 2.18)

1958-59 2.39) 2.32 2.95) 2.89

1959-60 2.37) 2.89)

1960-61 2.39) 3.14)

1961-62 2.02) 2.63)

1962-63 2.44) 2.87)

1963-64 2.51) 2.51 3.18) 3.29

1964-65 2.39) 3.49)

1965-66 3.18) 4.28)

1966-67 3.78) ) 4.44) )1967-68 3.07) 3.22) 3.77) 3.99)

1968-69 2.90) ) 3.75) )) )

1969-70 3.82) ) 3.50 4.18) ) 4.35

1970-71 3.74) ) 4.64) )1971-72 3.49) 3.79) 4.47) 4.57)

1972-73 4.10) ) 4.95) )1973-74 3.80) ) 4.59) )

Average(1950-51/1973-74) 2.80 2.98

Page 10: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

Ratios of Rural Saving and Private Investmentto Rural Income, India, 1950-51/1973-74

(1960-61 prices)6

S - Rural savingYr -Rural income

ndcfa - Private agricultural investment

.+*- *. ... *-, nd~~~~_Udcf a1 *

4 A

s/yr

3

2

1 **Ip S

1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974

Financial year beginning April

World Bank-20517

Page 11: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

expenditure data collected by the NSS. 1/ This population cannot afford any

positive saving. The capacity to save should also be negligible for millions

of people just above the poverty line. For as we shall see in the following

Section, saving bcomes significant only at levels of income much above the

poverty line. Thus only a fraction of the rural population are able to save.

And the saving ratio is significantly high for only a small minority of the

population. It is not surprising, therefore, that the average saving rate of

the rural economy as a whole should remain small even after some acceleration

of income growth in the last decade.

A comparison of trends in rural saving and rural investment ratios

cannot be properly made because the time series of rural investment is not

available. We have data only on private net capital formation in agriculture

(NDCFA) which excludes capital formation in rural non-agricultural sectors.

The ratio of NDCFA to rural income (YR) will:therefore be an underestimate of

the true rural investment/rural income ratio. But we find that over the whole

period of 24 years even this lower-limit ratio (NDCFA/YR) has averaged 2.98

percent which is slightly higher than the average rural saving ratio (2.80

percent).

The fact that the investment ratio is slightly higher than the

saving ratio is significant, for it implies that the urban sector has been

making a net contribution to capital formation in agriculture. In spite

of this contribution, however, the investment ratio exceeds the saving

ratio only by a small margin.

1/ Ministry of Planning reply to a question in Parliament, 29 June 1977.

Page 12: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

The contribution of the urban sector also explains the fact that,

although the rural household saving growth rate (3.5 percent) has remained

below the all-India household saving growth rate (5.6 percent), the agri-

cultural investment growth rate (7.0 percent) has exceeded the all-India

investment growth rate (4.6 percent). As a result the share of net capital

formation in agriculture in net domestic capital formation has gone up from

11.5 percent to 19.7 percent between 1950-51 and 1973-74. The proportion

still remains low, but there is no basis for the common impression that the

share of rural areas in aggregate annual investment has been falling.

(Table 2.)

TABLE 2

GROWTH RATES OF SAVING & INVESTMENT /a

1950-51/1973-74

Growth Rate

S1. Items Rural All-India

No. (At 1960-61 prices) TPercent)

1. Per Capita Household Saving 3.55 5.60

2. Per Capita Capital Formation /b 7.03 4.61

/a Saving in current prices is deflated by the implicit nationalincome deflator, and NDCFA in current prices is deflated by the

investment cost index. (See Appendix B.)

/b Agriculture only.

In spite of considerable investment in modern agricultural assets

the capital-output ratio in agriculture remained almost unchanged. It was

1.27 in 1950 and 1.28 in 1971. (Datta Roy Choundhury 1977.)

3. Marginal Propensities to Save and Invest

The marginal propensity to save has been derived with the basic

linear regression:

(1) St a + b YR + u

Page 13: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 10 -

Here S is rural saving and YR rural income. This equation has been estimated

with S and YR measured as aggregates in current as well as in constant prices,

and as real quantities per capita. (Table 3.)

The results show that the marginal propernsity to save 1/ (MPS) has

been 11 percent (Table 3, Equation 3.4), which is substantially higher than

the average (2.80 percent). Therefore, the elasticity 2/ of saving with

respect to income is quite high (about 3.93).

The marginal and average net investment rates 3/ turn out to be

18.2 and 2.98 percent and the elasticity 2/ of NDCFA with respect to rural

income is also high (5.88). (Table 4, Equation 4.4).

These are promising results for they imply that real rural saving

per capita can grow nearly four times and agricultural capital formation per

capita can grow nearly six times the growth rate of real rural income per

capita. If government policy is effective in sustaining a satisfactory rate

of growth of rural income, rural saving and farm investment can be relied

upon to grow much more than income. But of course the spontaneous growth

of rural saving and investment can be accelerated still further if banking

facilities are expanded in the rural area, and more attractive physical and

financial assets are offered.

Although the estimated income elasticities of rural saving and farm

investment are encouraging, the fact remains that the average saving and invest-

ment rates (about 3 percent) are disconcertingly low. Until the incidence of

1/ Estimated with real quantities per capita.

2/ At the means of variables.

3/ Estimated with real quantities per capita.

Page 14: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

TABLE 3

RURAL SAVINGS REGRESSIONS: INDIA

1950-51/1973-74

Serial Dependent Regression Coefficients -2

Number Variable Constant YR YR/N YR/NNPD YR/NNPD/N R Elasticity

3.1 S -174.658 0.043 0.974 1.392(7.438) (29.412)

3.2 S/N -5.984 0.047 0.967 1.562(8.622) (25.922)

3.3 S/NNPD -308.082 0.059 0.888 2.049(-6.821) (13.543)

3.4 S/NNPD/N -22.134 0.110 0.695 3.925(-5.442) (7.309)

S : Rural Saving in Current Prices N : Rural Population

YR : Rural Income in Current Prices NNPD : Net National Product Deflator

t-values are shown in the parentheses.

All equations are linear.

Page 15: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 12 -

poverty diminishes and income rises to a critical level, internal rural invest-

ment effort will have to be supplemented by resource flows from the urban non-

farm sector.

The sfving equation estimated by us enables us to compute the

"critical minimum" level of per capita income at which positive saving is

possible. The level is given by the horizontal intercept of the equation on

the income-axis (calculated by equating the estimated regression to zero).

Thus equation 3.4 yields Rs. 201.22 per annum as the critical per capita

rural income (in 1960-61 prices).

This figure may be compared to the Dandekar-Rath poverty line for

rural households: Rs 180 per annum in 1960-61 prices. Though it would have

been more appropriate to compare the poverty line with the intercept of a

cross-section regression, we can still draw the general inference that, on

the average, households in rural India can have positive saving only after

they have crossed the poverty line. The estimated intercept suggests that

real income per capita should rise at least 10 percent above the poverty line

before a typical household can afford to save.

We can even conceive of using the critical minimum income, esti-

mated from a saving regression, as a legitimate alternative way of defining

the poverty line itself, the logic being that a household may be properly

supposed to have risen above poverty only if and when its income exceeds

consumption.

There is another interesting question on which the estimated rural

saving equation throws some light. We can ask: how much will rural income

have to rise in order that higher rural saving ratios of the order of, say

5, 8 and 10 pereent may be achieved? This question can be answered if we

Page 16: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

TABLE 14

AGRICULTURAL INVESTMENT REGRESSIONS: INDIA1950-51/1973-74

Serial Dependent Regression Coefficient -2Number Variable Constant YR YR/N YR/NNPD YR/NNPD/N R Elasticity

4.1 NDCFA -217.693 0.048 0.966 1.525(7.741) (25.104)

4.2 NDCFA/N -7.758 0.053 0.930 1.752(6.876) (17.059) I

4.3 NDCFA/IDW -556.200 0.088 0.932 2.717(10.841) (17.453)

4.4 NDCFA/IDW/N -40.480 0.182 0.689 5.876(5.848) (7.060)

NDCFA : Net Domestic Capital Formation in Agriculture N : Rural Population

YR : Rural Income in Current Prices NNPD : Net National Product Deflator

t-values are shown in the parentheses IDW : Rural Investment Deflator (Weighted)

All equations are linear.

Page 17: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 14 -

rewrite the estimated saving function in the form

(2) S/Y = 0.11 - 22.13y

This saving ratio function is pictured in Figure 3. In the Fourth Plan period

(1969-70/1973-74) the saving rate had risen to 3.8 percent while per capita

rural income averaged Rs 291.95 (in 1960-61 prices). The equation suggests

that a 5 percent saving rate would require income to rise to about Rs 369;

an 8 percent saving rate would require an income of Rs 738 and a 10 percent

rate Rs 2,213. In other words, the Fourth Plan level of rural income would

have to increase 2.5 times and 7.5 times for the rural saving rate to become

8 and 10 percent respectively. And even a 5 percent saving rate would require

a 26 percent increase in rural income. Since the rate of growth of rural in-

come per capita has averaged only 0.8 percent per annum over a long period,

it is clear that a substantial escalation of the rural income growth rate

will be necessary before the rural saving rate attains a reasonable level and

the rural growth process becomes self-sustained. Meanwhile, a net inflow

of resources from urban to the rural sector will continue to be necessary.

4. Propensities in the Fifties and Sixties

As stated in the beginning of Section 2, the foregoing results

have been derived with a 24-year time series of rural saving constructed by

pooling the Reserve Bank figures for the first 10 years and the CSO figures

for the latter 14 years. The pooled series has enabled us to get long-

period magnitudes of saving propensities. But there are some conceptual

differences between the RBI and CSO series. And there is reason to believe

that the saving propensity should have been higher in the second period which

Page 18: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

Average Rural Saving FunctionIndia, 1952-53/1973-74

0.120... . . . . . .. . . . . . .. . . . . . . . . . . . ................. ... . . . . . . .........-.------1......----1S--"-----

0.10-

~~~~~~~~~~~~~~~~~~~~~~~S0.0 =011 -22. 13 y-10.08

0.06- / .1ll

0.04__ _ _ _ _ _

0.02

0 1 200 600 1000 1400 1800 2200 Y

-0.02-

s = Average saving rate-0.04 - r = Average saving rate out of ,er capita real income (per cent)

y = Per capita income in 1960-61 prices (Rs) (Rs.)

-0.06 anl01

World Bank-20518

Page 19: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 16 -

includes the years of the Green Revolution. Therefore, separate regressions

have also been estimated for the two periods. The key results are summarized

in Table 5.

TABLE 5

RURAL SAVING PROPENSITIES IN THE FIFTIES AND SIXTIES

Marginal AveragePeriod (Percent) (Percent) Elasticity

I 1950-51/1962-63 1.33 2.33 0.57

II 1960-61/1973-74 11.79 3.14 3.75

I + II 1950-51/1973-74 11.03 2.80 3.93

It is evident that the marginal propensity to save (MPS) (out of

real income per capita) in the second period is about 9 times what it was in

the first period. Some of this large difference is perhaps due to the under-

estimation of saving in the earlier period by the Reserve Bank. But even

allowing for this, it is clear that with the more rapid growth of rural income

per capita in the second period the incremental saving rate escalated 1/. And

the elasticity of saving with respect to rural income which was slightly more

than half in the early period increased to nearly 4 in the latter period. How-

ever, the average saving rate went up only moderately from 2.3 to 3.1 percent.

Our estimates of saving propensities for the earlier period are

comparable with those derived in an earlier study by Uma Datta Roy Choudhury

(1968). She estimated the marginal propensity to save out of per capita real

income in rural India to be 0.96 percent over the period 1950-51/1962-63. Our

estimate is 1.33 percent. Her estimate of the average propensity to save is

1/ Growth rates of real rural income per capita during 1950-51/1962-63and 1960-61/1973-74 were 0.42 percent and 1.17 percent per annumrespectively.

Page 20: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 17 -

2.06 percent and ours 2.33 percent (Table 6). The difference is mainly due

to the different concepts of income used 1/. But in both studies the co-

efficient of determination of the saving function for the earlier period is

very low (0.11, 0.14) and the estimated marginal propensities to save are not

significant 2/.

TABLE 6

RURAL SAVING PROPENSITIES OBTAINED IN VARIOUS STUDIES: INDIA(1950-51/1962-63)

MPS* APS* Elasticity Deflator Deflator

(Percent) R2 of Saving of Saving of Income

Uma DattaRoy 0.96 2.06 0.14 0.14 Construc- GNP

Choudhury /b (1.32) /a tion Cost Deflatorof RuralBuildings

K. L. Gupta /c 3.05 - 0.91 - GNP GNP

(10.80) Deflator Deflator

Krishna andRaychaudhuri /b 1.33 2.33 0.11 0.57 NNP NNP

(1.14) Deflator Deflator

/a Parentheses show t-value./b Real capita rural saving out of real per capita rural income.

/c Real per capita rural saving as a function of real per capita dis-

posable agricultural income.* MPS = Marginal propensity to save. APS = Average propensity to save.

1/ We have applied the NSS rural/urban consumption rato to CSO estimates

of aggregate consumption to get rural consumption. She has applied the

same ratio to aggregate national income to get rural income.

2/ There is one more study by K. L. Gupta, (1970), in which a much higher

and significant estimate of the marginal propensity to save (3.05 per-

cent) was obtained for the same period. (Table 6.) His results are

surprisingly different from and inconsistent with those of our study

and the Datta Roy Choudhury study. The saving data used are the same

in all the three studies. But income is defined by Gupta as real, dis-

posable agricutural income per capita, the deflator being the implicit

GNP deflator. Unfortunately his income series is not available for

comparison.

Page 21: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 18 -

A few all-India cross-section studies of rural saving have also

been completed with the National Council of Applied Economic Research (NCAER)

data for 1962, 1967-68 and 1968-71. The average and marginal propensities to

save estimated in these studies are summarized in Table 7. The MPS is avail-

able only for two years: 1962 which is the last year of our earlier period

(1950-51/1962-63), and 1967-68 which lies in the middle of our second period

(1960-61/1973-74) and was the first bumper crop year after the introduction

of the new hybrid-seed technology. The saving propensities in Tables 5 and 7

TABLE 7

ESTIMATES OF RURAL SAVING PROPENSITIES

Household Sample MPS* APS* 2

Year Size (Percent) R Elasticity

NCAER Data (Cross-Section)

1962 8,527 16.8 5.5 0.93 3.05

1967-68 2,380 36.0 8.2 - 4.39

1968-69 /a 4,118 - 3.0 - -

1969-70 Ia 4,118 - 5.0 - -

1970-71 /a 4,118 - 5.9 - -

1970-71 /b 4,118 22.0 6.9 0.29 -

Time SeriesEstimates(This Paper) Period II 11.8 3.1 0.60 3.75

/a MPS were not estimated with data from these years.

/b Bhalla (1976). Measured Income.* MPS = Marginal propensity to save.

APS = Average propensity to save.

are not strictly comparable as they are derived from time series and cross-

section equations respectively. But it does seem that cross-section estimates

of saving propensities, regarded as long-period equilibrium parameters, tend

to be higher than time-series estimates. It is also clear that the cross-

section average saving rate fluctuates from year to year and, as we would

Page 22: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 19 -

expect, it is higher in the good crop seasons such as 1967-68 and 1970-71

than in bad or average seasons.

In a recent study Bhalla (1976) has estimated alternative cross-

section saving equations with 1970-71 NCAER data. He defines saving as change

in net worth. Income is defined in three alternative ways: measured income

(Y) and two concepts of permanent income (Y and Y ). Y is estimatedp35 px px

by adding up the returns to all household assets. Y 35 is estimated with

particular weights attached to the measured income of the current and two

preceding years. Transitory consumption occurs as an additional variable

in the Bhalla saving function besides the income variables. The marginal

propensity to save with respect to measured and permanent income turned out

to be 22 and 23 percent respectively. The APS was 6.9 percent with the net

worth concept of saving. Both these cross-section estimates are also higher

than our time series estimates.

5. Saving Out of Permanent and Transient Income

In another set of experiments an attempt has been made to differen-

tiate between transient and permanent income as contributors to saving. Per-

manent income in year 't' is defined as the average of measured income in the

current and two preceding years (t, t-l and t-2). Thus:

(3) YPt ( t t- + t-2 )/3.

Transient income in period 't' is then the difference between measured income

and permanent income:

(4) YTt = yt ( t t-l t-2)/3-

The saving function is

(5) St a + b YPt + c YTt + ut.

Page 23: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 20 -

The estimated function (5) with data for 24 years 1950-51/1973-74 in per

capita real terms is

(5') St = - 29.587 + 0.139 YP, - 0.026 YTt*(-9.113) (11.518) (-0.884)

(-2 = 0.862)

(t-values are shown in the parentheses.)

The implicit elasticity of saving with respect to permanent income (at means

of variables) is 4.85.

The coefficient of transient income is non-significant. But the

marginal propensity to save out of permanent income turns out to be higher

(13.9 percent) than the MPS out of measured income (11.0 percent). (Regres-

sion 3.4 in Table 3.)

This implies that, if the growth of income was more stable, the pro-

pensity to save would be higher. But because income fluctuates widely from

year to year saving is also unstable and the propensity to save is reduced.

In order to study the effect of income fluctuations on saving we have com-

puted two components of measured saving: (1) saving out of permanent income

which may be called "permanent saving" and (2) "transient saving" computed

as measured saving minus permanent saving. These components are shown in

Table 8 and Figure 4. It is interesting to observe that in 6 out of 22 years

irregular saving out of windfall income augmented permanent saving by 8.2

percent on the average. But in the remaining 16 years windfall losses reduced

permanent saving by 9.2 percent on the average, the reduction ranging between

0.5 percent and 22.9 percent. Thus, income fluctuations produce measureable

variations in saving, from year to year, of the order of 8 to 9 percent on

the average above or below the normal level.

Page 24: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 21 -

TABLE 8

DECOMPOSITION OF MEASURED SAVING INTO "PERMANENT" AND "TRANSIENT"SAVING: INDIA (1952-53/1973-74)

"Measured" "Permanent" "Transient" Ratio of TransientYear Saving Saving Saving to Permanent Saving

(Rs Per Capita in 1960-61 Prices) (Percent)

1952-53 5.69 5.46 0.23 4.21

1953-54 6.12 5.99 0.13 2.17

1954-55 5.50 5.93 -0.43 -7.251955-56 5.50 5.73 -0.23 -4.011956-57 6.10 5.42 0.68 12.551957-58 5.62 6.76 -1.14 -16.861958-59 6.30 7.37 -1.07 -14.521959-60 6.12 7.65 -1.53 -20.00

1960-61 6.26 6.95 -0.69 -9.931961-62 5.23 6.78 -1.55 -22.861962-63 6.28 6.74 -0.46 -6.821963-64 6.47 6.56 -0.09 -1.37

1964-65 6.66 7.44 -0.78 -10.481965-66 8.60 8.04 0.56 6.971966-67 10.22 8.63 1.59 18.421967-68 8.92 9.21 -0.29 -3.151968-69 8.47 10.21 -1.74 -17.041969-70 11.05 11.11 -0.06 -0.541970-71 10.92 11.17 -0.25 -2.241971-72 10.31 11.33 -1.02 -9.00

1972-73 11.84 11.28 0.56 4.961973-74 11.20 11.39 -0.19 -1.67

Maximum 11.84 11.39 1.59 (a) +18.4(b) - 0.5

Minimum 5.23 5.42 -1.74 (a) + 2.2(b) -22.9

Mean 7.70 8.05 -0.35 (a) + 8.21(b) - 9.23

(a) For positive terms.

(b) For negative terms.

Page 25: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

Per Capita Rural Saving: India, 1952-53/1973-74

Rs. (1960-61 prices)1 2

12 ~~~~sr - Measured saving

sp -Permanent savin-g. 10- st - Transient saving / /

6 ~~~~~~~~~~~~~~~~~~sr

4

2 S4~~~~~~~~~~~~~~ .

-2 -~~~~~

O ,, * ,

-2~~9 0 ' *.,., 99- ". .,.- 9 9 9 9 9 - --. * 9'--*.,..,

195;2 1954 1956 1958 1960 1962 194 1966 1968 1970 1972 1974

Financial year beginning April

World Bank-20519

Page 26: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 23 -

6. Some Disaggregative Estimates

The analysis presented here is basically "Keynesian" and

"aggregative" in the sense that income is the sole variable used to explain

saving, and the argument abstracts from variations in saving rates across

different classes of farmers, and different regions of the country. Ideally,

the influence of the varying returns to investment on the rates of saving

of various groups of farmers, in various regions, should be analyzed. But

disaggregated data on returns, and regional rural income and saving series,

are not available.

However, there are bits of evidence suggesting (1) that the rural

saving rate has been relatively higher in the irrigated, new technology

regions; (2) that, therefore, the increase in the saving rate in the late

sixties and early seventies should have been partly induced by the higher

returns expected from new technology; (3) that returns from working capital

as well as fixed investment in new technology have indeed been quite attrac-

tive; and (4) that the composition of rural capital has been changing pro-

gressively in favor of assets embodying more productive, modern technology.

Substantial data on this last point are presented in Section 7 below.

On the saving being higher in more irrigated progressive areas, we have the

results of some local cross-section studies. For the famous high-technology

district of Ludhiana in Punjab State the average saving rate in the first

five years of the introduction of HY seed varieties (1966-67/1970-71) was

found to range between 22 and 35 percent on a sample of 72 farms. (In this

district the irrigation ratio rose to 80 percent and the HYV area ratio to

nearly 60 percent in those years.) (Kahlon et.al. 1972). In the Nainital

district of U.P. State in 1967-68 the average investment/income ratio on a

Page 27: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 24 -

sample of 156 farms is reported to have been as high as 17 to 44 percent.

And it was distinctly higher on innovating small and medium farms.

INVESTMENT RATIO ON A SAMPLE OF NAINITAL FARMS, 1967-68

Investment Ratio Irrigation Ratio HYV Area Ratio

(Percent) (Percent) (Percent)

Innovating Farms:Small 28.14 74.0 28.1

Medium 44.3 75.0 24.1Large 43.3 78.8 31.8

Other Farms:Small 26.7 38.0 4.5

Medium 17.5 37.2 4.0

Large 43.7 48.7 6.2

Source: Shah and Singh 1969.

Similar evidence is available about some districts of South India. In a

sample of 36 farms the average investment ratio was 18 percent in Baptala

district and 31 percent in Chittoor district in Andhra Pradesh State in

1965-67. (The irrigation ratio was 30 percent in Baptala and 42 percent in

Chittoor.) (Misra and Mallick 1969). In the well-endowed district of

Coimbatore in Tamil Nadu State of South India the saving rate on a sample of

30 farms (with a 40 percent irrigation ratio) averaged 10 percent on small

and medium farms and 17.8 percent on large farms. (Rajagopalan and

Krishnamoorthy 1969.) For Western India a study of 87 Sangli farms in

Maharashtra State estimated saving rates averaging 12 to 16.6 percent for

four subgroups in 1970-71. (Chauhan et. al. 1972).

These estimates of saving/investment rates are not strictly com-

parable because different concepts of income, saving and investment have

been used by the authors. In general, they are gross magnitudes, whereas

our all-India time-series pertain to net quantities. However, the general

Page 28: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 25 -

level of local, cross-section saving/investment ratios varying between 10

and 44 percent, is clearly far higher than (3 to 15 times) the all-India time

series ratios of the order of 3 percent (Table 1). Therefore, there can be

little doubt that in the better-irrigated, more rapidly innovating regions,

saving/investment rates tend to be much higher than in the country as a whole.

In part, the better irrigation/innovation situation should be influencing

saving by raising the level of income itself. But to some extent it may also

be inducing farmers directly to save/borrow/invest proportionately more at the

margin by raising expected returns. Unfortunately, the lack of data does not

allow the separation of these effects.

In spite of high saving/investment rates in some regions, the all-

India saving/investment rate remains low because the poorer, drier and less

dynamic regions (farms) dominate it. To cite an example, the average (net)

saving rate was found to be only 2.2 percent for a set of 72 farms in the

Hooghly district of West Bengal State in 1970. (Chakravarti 1970). This is

even less than the all-India average.

Just as the variation of the technological level underlies wide

cross-section differences in the saving rate, it is also the basic explana-

tion of the rise in the saving rate over time after the mid-sixties (already

noted in Section 4.)

As regards rates of return to modern input and assets it may be

noted, first, that physical yield increases of the order of 100 to 200 percent

above present levels are feasible with the modern seed-water-fertilizer

technology. (Singh 1978.) Cereal yields can be raised from less than 1,000

kgs. per hectare to 2,500 kgs. in the case of rice, and 4,000 to 5,000 kgs.

in the case of wheat. And pulses and oilseed yields can be raised from

Page 29: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 26 -

600-700 kgs. per hectare to 1,000-1,500 kgs. per hectare. Second, under

irrigated conditions, marginal physical output-input ratios of the order of

10 or more, for fertilizer nutrient inputs, are quite common. Financial

return-cost ratios for fertilizer use lie between 2 and 3 for paddy and 2.9

to 4 for wheat. (FAI 1978). These facts make investments in modern seed-

water-fertilizer technology very attractive. When medium and long-term

investments are considered, the return prospects are equally attractive.

On the basis of a number of studies of returns to net farm investments, the

Agricultural Refinance and Development Corporation of India (ARDC) has found

that the financial return to minor irrigation investments ranges between 20

and 50 percent. The return to land development in command areas of irriga-

tion systems exceeds 50 percent. The return to dairy-buffalo units also

exceeds 50 percent. The return to mechanised fishing boats ranges between

20 and 25 percent and the return to investment in citrus, coconut, pepper,

cashew and rubber investments ranges between 18 and 49 percent. The social

return is of course much higher in all these cases. (ARDC 1978.)

These figures point to the existence of strong incentives for the

raltively higher rural saving/investment rates in the more dynamic, irrigated

regions, and in the recent decade of accelerated technical change.

7. Tangible Wealth and Asset Composition

In a recent paper Datta Roy Choudhury (1977) has presented revised

and comparable estimates of reproducible tangible wealth (RTW) in various

sectors of the Indian economy in 1950, 1961, 1966 and 1971. Her concept of RTW

in agriculture includes farm buildings, construction works, land improvement,

irrigation structures, livestock, machinery, transport and other equipment,

Page 30: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 27 -

and inventories. (The value of land and intangibles is excluded.) Her

estimates show that in constant (1960-61) prices the RTW in agriculture in-

creased from Rs 6,514 crores in 1950 to Rs 10,625 crores in 1971 at the

rate of 2.36 percent per annum. But the share of agriculture in the total

national RTW declined from 30 percent to 18 percent because the growth of

total RTW was faster (4.9 percent per annum). Thus it is clear that the

growth of RTW has been much slower in agriculture than in the whole economy.

This findings is not inconsistent with the fact noted in Section 2

that net capital formation in agriculture has been growing at a higher rate

(7.0 percent) than net capital formation in the whole economy (4.6 percent).

The explanation is that these latter figures refer to the growth of annual

investment and the figures in the preceding paragraph to the growth of the

total stock. It is quite possible that the share of agriculture in annual

investment increases and yet its share in the total stock falls if its initial

share in investment as well as in stock is very low.

As the figures in Table 9 show, this is, in fact, what really

happened. While the share of agriculture in the increment of RTW increased

from 7 percent in the decade 1951-61 to 15 percent in the decade 1961-71,

its share in the total stock of RTW fell steadily from 30 percent to 20 per-

cent in the first decade and 20 percent to 18 percent in the second decade.

Page 31: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 28 -

TABLE 9

SHARE OF AGRICULTURE IN REPRODUCIBLE TANGIBLE WEALTH (RTW)

AND IN THE INCREMENT OF RTW, 1950-51/1970-71

Decade DecadeIncrease Increase

RTW RTW in RTW in RTWAgriculture Total Agriculture Total

(Rupees Crores* in 1960-61 Prices)

1950-51 6,514 21,485 - -

(30.3)1960-61 7,685 38,990 1,171 17,505

(19.7) (6.7)

1970-71 10,625 58,546 2,940 19,556(18.1) (15.0)

Figures in parentheses are percentage shares in the total.

*1 crore = 10 million.

Source: Datta Roy Choudhury (1977)

Another noteworthy finding of a study by Divatia (1976) is that

the concentration ratio for rural assets remained undiminished between 1961-62

and 1971-72. It was 0.65 in 1961-62 and 0.66 in 1971-72. The bottom 10 per-

cent of rural households owned 0.1 percent of total rural assets, and the top

10 percent owned 51 percent of rural assets both in 1961-62 and 1971-72.

This has the important implication that up to 1971-72 the land reform legis-

lation and other policies had no impact on the distribution of rural property.

In order to study the composition of assets (other than land) we

have calculated the aggregate value of the following aricultural assets in

constant 1960-61 prices: wooden and iron ploughs, carts, sugarcane crushers,

persian wheels, fishery equipment, electric and oil pump sets, tractors, oil-

crushers, other modern equipment, livestock, land improvement and irrigation

Page 32: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 29 -

works, and residential and non-residential house property. 1/ (Table 10)

Figures for land improvement, irrigation works, house property and fishing

equipment have been obtained from Reserve Bank data and for other items from

the quinquennial Livestock Census data. The total value of these assets ex-

cluding residential house property (in 1960-61 prices) increased from 10,285

crores to Rs 15,591 crores, i.e., by 81 percent over the 21-year period 1951

to 1972. The implicit annual growth rate is 1.97 percent. 2/

When we examine trends in the composition of assets, we find that

livestock accounted for the largest share (66 percent) in the total value of

assets in 1951. But this share declined to 53 percent in 1972.

The second largest component of RTW, namely, land improvement and

irrigation works, accounted for 17 percent of the total reproducible tangible

wealth in 1951. 3/ But this proportion rose to nearly 25 percent in 1972.

1/ Some modern agricultural implements, viz., shellers, harvester combines,planters, sprayers and dusters have not been included in our RTW esti-mates for 1972 as data on their prices could not be obtained.

2/ The items covered and the sources of quantity and price data used by usare given in Appendix C. Our estimate of the total value of assets ishigher than the Datta Roy Choudhury (1977) estimate. (Table 11). Thedifference is due to four factors. First, we have valued the physicalstock in every Livestock Census year at the same price without allowingfor any depreciation of the depreciated part of the stock. Secondly,the prices used by us may be different from those used by Datta RoyChoudhury in the case of some assets. Our price sources are given inAppendix C; the Datta Roy Choudhury sources are not known to us. Third-ly, the price deflation procedures used by us may be different fromDatta Roy Choudhury's. Fourthly, our estimates include private invest-ment only. This makes a difference particularly in the case of landimprovement and irrigation works.

Our reason for developing independent estimates was the intention tostudy the changes in the composition of assets. The Datta Roy Choudhurystudy did not detail the composition.

3/ Public investment in land improvement and irrigation works, for 1949-50,1960-61 and 1965-66, amounting to Rs 229 crores, Rs 1,158 crores andRs 2,300 crores respectively (in current prices) has not been includedhere.

Page 33: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 30 -

The share of non-residential farm buildings declined from 12 to 9

percent.

TABLE 11

REPRODUCIBLE TANGIBLE WEALTH IN AGRICULTURE /a

Annual CompoundValue Growth Rate

(Rs Crores 1960-61 Prices) (Percent)

Uma Datta 1950 1961 1971RoyChoudhury /b 6,514 7,685 10,625 2.36

Raj Krishna 1951 1961 1972andRaychaudhuri /c 10,285 12,124 15,591 1.97

/a Excluding residential house property.lb Including public sector investment.Ic Excluding public sector investment.

But the share of machinery and equipment increased. In fact, it escalated from

5 percent to 13 percent during 1951-72.

Thus a significant modernization of the asset structure seems to have

occurred, with livestock growing at the low rate of 1 percent, non-residential

building stock too at the low rate of 0.6 percent, but land improvement and

irrigation facilities growing at the rate of 3.8 percent and machinery and

equipment more than 6 percent per annum for two decades.

In Table 12 we have analysed the growth in the stock of agricultural

equipment at a disaggregated level. It appears that two traditional items,

wooden ploughs and carts, accounted for 78 percent of the total value of equip-

ment in 1951. But their share fell to 43 percent in 1972. Thus the relative

importance of these traditional items in the stock of equipment declined

during the period. The number of wooden ploughs increased only 25 percent,

and carts only 33 percent; but on the other hand the number of improved iron

Page 34: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 31 -

ploughs increased nearly 6 times, from less than 1 million to more than 5

million.

A similar trend is discernible in the case of irrigation equipment.

The number of traditional persian wheels increased only marginally by 4 per-

cent over the whole 21-year period. Their share in the total value of equip-

ment declined from 5.5 percent to 1.5 percent. On the other hand, the number

of oil and electric pumps increased more than 28 times, from about 96 thousand

in 1951 to nearly 2.8 million in 1972. In fact, in 1972 pumps became the

second largest component of farm equipment accounting for 30 percent of the

total value of equipment, next to carts accounting for 40 percent. Here we

have clear evidence of heavy private investment in irrigation during the two

decades.

The third most important component of farm equipment came to be

tractors. Their number grew more than 16 times from about 7,800 to 125,000.

And their share in the total value of equipment rose from 2 percent to 12

percent.

The investment in bullock-driven cane-crushers declined from Rs 30

crores to Rs 27 crores, while the investment in power crushers increased three

times, from Rs 4 crores to Rs 12 crores (in 1960-61 prices).

The investment in traditional oil crushers also declined from

about Rs 11 crores to Rs 4 crores. But the investment in fishery equipment

increased phenomenally: more than 8 times from Rs 15 crores to Rs 117 crores.

The overall picture that appears, then, is that the total value of

equipment increased 3.8 times from Rs 544 crores to Rs 2,072 crores; and the

composition of equipment underwent a radical transformation, with the share

Page 35: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 32 -

of traditional items, e.g., carts, wooden ploughs, persian wheels, bullock-

driven cane-crushers and very small oil-crushers declining, and the relative

share of modern equipment including iron ploughs, oil and electric pumps,

tractors, power crushers and fishery equipment increasing.

Thus there can be no doubt that Indian farmers, especially medium

and large farmers, have been modernizing their asset structure to accommodate

more productive forms embodying the new technology. The constraints on the

growth of productive investment are obviously not due to any lack of adaptive-

ness on their part, but to the low overall saving rate, inadequacy of external

finance, lack of knowledge of profitable investment opportunities, and supply

bottlenecks.

Page 36: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 33 -

TABLE 13

RURAL SAVING, INCOME AND CAPITAL FORMATION, INDIA1950-51/1973-74

Private Invest-ment in

Rural Saving Agriculture Rural Income

(Rs Crores)

1950-51 166.30 59.35 7,088.60

1951-52 170.70 99.90 7,407.93

1952-53 163.50 57.15 7,306.39

1953-54 180.50 100.27 7,636.28

1954-55 147.90 104.84 6,672.79

1955-56 153.70 135.54 6,992.99

1956-57 187.70 233.36 7,763.85

1957-58 179.50 181.71 8,843.81

1958-59 212.20 246.10 8,870.73

1959-60 212.50 248.82 8,959.99

1960-61 225.30 295.65 9,422.11

1961-62 195.80 264.77 9,711.98

1962-63 248.30 303.45 10,195.37

1963-64 284.30 348.14 11,314.87

1964-65 326.80 434.01 13,668.12

1965-66 467.80 559.47 14,706.57

1966-67 649.50 630.07 17,170.65

1967-68 628.60 622.96 20,500.91

1968-69 603.00 672.26 20,790.97

1969-70 834.25 820.46 21,851.08

1970-71 864.75 967.56 23,131.05

1971-72 868.60 933.59 24,892.58

1972-73 1,125.00 1,152.09 27,400.011973-74 1,279.27 1,406.76 33,632.43

Annual CompoundGrowth Rate(Percent) 10.15 13.92 7.51

For sources and methods of estimation, see Appendices A and B.

Page 37: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 34 -

APPENDIX A

THE RURAL HOUSEHOLD SAVING SERIES 1950-51/1973-74

Estimates of net rural household saving in India for the period

1950-51 to 1962-63 are available in the Reserve Bank fo India (RBI) Bulletin,

March 1965. The RBI estimate is based mainly on data collected in the Bank's

All India Rural Credit Survey of 1951-52 (AIRCS), the Follow-Up Survey of

1956-57 and the Rural Debt and Investment Survey of 1961-62. Saving in the

form of physical assets and inventories was estimated with these bodies of

data for the three bench-mark years 1951-52, 1956-57 and 1961-62.

For the estimate of financial investments of the rural sector the

Reserve Bank used its own survey data as well as data collected in the Rural

Household Saving Survey (1962) by the National Council of Applied Economic

Research (NCAER). Adding financial investments to estimated saving in the

form of physical assets and deducting net borrowings from the urban sector

the Reserve Bank computes the net saving of the rural household sector.

The estimated saving for the three bench-mark years formed 3.3 per-

cent, 3.7 percent and 3.3 percent of agricultural income (including income

from animal husbandry, forestry and fishery). The average of these three

ratios, 3.4 percent, was applied by the Reserve Bank to the time series of

agricultural income to get the time series of saving for the whole period

1950-51/1962-63 because the variation in thle average saving as between bench-

march years was found to be very small.

For the period 1960-61 onwards the Central Statistical Organization

(CSO) has published estimates of net saving in the entire national household

sector in the National Accounts Statistics, October 1976. The breakup of total

household sector saving into rural and urban saving has not been given by the

Page 38: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 35 -

CSO. But since for the earlier period (1950-51/1962-63) Reserve Bank figures

for total as well as household saving are available, the time series of the

ratio of rural to total saving for this period was computed. The average for

the nine years 1954-55/1962-63 turned out to be 25 percent with a very small

coefficient of variation. Hence we have computed the rural saving series for

the recent period 1960-61/1973-74 by applying a constant ratio (25 percent)

to the total household saving series.

The combined time series of rural saving for the whole period

1950-51/1973-74 includes figures for 1950-51 to 1959-60 based on RBI data and

figures calculated for 1960-61/1973-74 on the basis of CSO data.

An alternative rural saving series for the later period 1960-61/

1973-74 was also generated without using a constant ratio between rural saving

and total household saving. Rural saving (RBI) was regressed on total house-

hold saving (RBI) for the period 1954-55/1962-63 and the estimated regression

was used to estimate rural saving (1960-61/1973-74) with the given CSO series

of total household saving. The two alternative rural saving series computed

by the "constant ratio method (CRM)" and the "regression method (RM)" are

shown in Table A.1. The RM estimates are consistently lower than CRM estimates,

the difference widening from Rs 4 crores in 1961-62 to Rs 248 crores in 1973-74.

The lower series also yields a lower marginal propensity to save for the whole

24-year peirod (6.9 percent) than the higher series (11.0 percent) used in

the text (equation 3.4).

Page 39: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 36 -

TABLE A.1

RURAL SAVING IN INDIA(1960-61/1973-74)

Year Saving CRM Saving RM CRM - RM(Rs Crores)

1960-61 225.30 215.02 10.281961-62 195.80 192.17 3.631962-63 248.80 233.23 15.571963-64 284.30 260.74 23.561964-65 326.80 293.66 33.141965-66 467.80 402.91 64.891966-67 649.50 543.72 105.781967-68 628.50 527.45 101.051968-69 603.00 507.69 95.311969-70 834.25 686.86 147.391970-71 864.75 710.49 154.261971-72 868.50 713.39 155.111972-73 1,125.00 912.12 212.881973-74 1,279.25 1,031.63 247.62

Annual CoumpoundGrowth Rate(Percent) 15.67 13.97

The main reason for preferring the higher (CRM) series for the

analysis in the text is that the rate of growth of the CRM series in constant

prices (5.6 percent) is closer to the rate of growth of net domestic private

capital formation in agriculture (7.0 percent) than the lower RM series

(4.4 percent) (Table A.2). It is also closer to the rate of growth of the

deposits of primary agricultural credit cooperatives (11.4 percent). The RM

series does seem to underestimate real rural saving.

Page 40: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 37 -

TABLE A.2

GROWTH RATES OF RURAL SAVING,INVESTMENT AND DEPOSITS

1950-51/1973-74(1960-61 PRICES)

Rural Saving (CRM) 5.61Rural Saving (RM) 4.38Net Domestic Private Capital

Formation in Agriculture 9.17Primary Agricultural Credit

Cooperative Deposits 11.37

An additional reason for using the CRM series is that for four years

for which independent cross-section data are available from the NCAER the

average saving ratio computed from the CRM series is nearer the cross-section

ratios. (Table A.3).

TABLE A.3

AVERAGE SAVING RATIOS

BENCHMARK YEARS

NCAERCross-Section CRM Series RM Series

Ratio Ratio RatioYear (Percent) (Percent) (Percent)

1961-62 5.5 2.02 1.981968-69 3.0 2.90 2.451969-70 5.0 3.82 3.161970-71 5.9 _ 3.74 3.09

Page 41: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 38 -

APPENDIX B

NET CAPITAL FORMATION IN AGRICULTURE SERIES1950-51/1973-74

Estimates of capital formation in the agricultural sector for the

period 1960-61/1972-73 are obtained directly from NAS, 1976 (National Accounts

Statistics 1960-61/1974-75, October 1976). They include capital formation in

agriculture proper, forestry and fishing, i.e., capital formation in the

primary sector minus capital formation in mining. For the earlier period

(1950-51/1959-60), estimates have been derived by extrapolating backwards the

share of the agricultural sector in total net capital formation.

Net capital formation in agriculture (NDCFA) in current prices has

been deflated to get NDCFA in 1960-61 prices. The deflator is the weighted

average of the real construction cost series prepared by R.N. Lal (1977) and

the machinery and equipment wholesale price series constructed by the authors.

The weights used are .9 and .1 respectively for the construction and equipment

indices. The weights are chosen on the basis of the average asset composition

of rural reproducible tangible wealth in 1951, 1961 and 1972.

Page 42: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 39 -

APPENDIX C

TANGIBLE WEALTH OF RURAL HOUSEHOLDS

The list of items included in our definition of tangible wealth

and the sources of quantity and price data used to evaluate them are shown

in the following table.

QUANTITY AND PRICE DATA USED FOR ESTIMATING TANGIBLE WEALTH

Source ofSource of Price Data/ Source of

Item Quantity Data Deflator Value

Machinery and Equip- Quinquennial CSO for some Reserve Bankment including items Livestock States. Min- of India (RBI)1 to 7 in Table 12. Census. nistry of Bulletin.Item 8 in Table 12 Agricultureincludes seed drills, for some items.threshers and tractors- Wholesale Priceoperated mould board Index forand disc ploughs, disc Machinery andharrows, tillers, Equipmentlevellers, seed-cum- deflation.fertilizer drillsand trailers.

Forestry and Fishery Quinquennial Wholesale Price RBI Bulletin.Equipment Livestock Index for

Census. Machinery andEquipment.

Livestock Quinquennial Agricultural RBI BulletinLivestock Situation in and AIDIS*,Census. India (Ministry RBI.

of Agriculture).CSO for someitems.

Rural Buildings R.N. Lal's Mukherjee(including Residen- Construction and Shastry,tial and Non- Cost Index for AIDIS, RBI.Residential). Deflation

Land Improvement R.N. Lal's Mukherjeeand Irrigation Construction and Shastry.Works. Cost Index RBI Bulletin.

Land RBI Bulletins.AIDIS, RBI.

* AIDIS: All-India Debt and Investment Survey (1971-72), RBI.

Page 43: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 40 -

Some items could not be included, namely, certain types of equip-

ment and plantations because comparable quantity data and/or price data for

them were not available. Durable consumer goods have also not been included.

Details of extrapolations/intrapolations and deflations used are available

with the authors.

Page 44: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 41 -

REFERENCES

1. ARDC 1978: Agricultural Refinance and Development Corporation,"Trends in Long Term Investment in Agriculture," Bombay, May 1978.

2. Bhalla, Surjit S.: Aspects of Savings Behaviour in Rural India,Princeton University and Brooking Institution, 1976. (Mimeo).

3. Central Statistical Organization: National Income Statistics -Estimates of Gross Capital Formation in India for 1948-49 to1960-61, New Delhi, 1961.

4. Central Statistical Organization: Estimates of National Income1948-49 to 1962-63, New Delhi, February 1964.

5. Central Statisticsl Organization: National Accounts Statistics1960-61 -- 1974-75, New Delhi, October 1976.

6. Chakravarty, S.K., "A Recent Change in Saving - Investment Directionof the Small Cultivators in West Bengal (Case Studies of HooghlyDistrict)," Indian Journal of Agricultural Economics, Vol. XXVII,October-December 1972, pp. 64-74.

7. Chauhan, K.K.S., S. Mundle and D. Jadhav, "Income, Savings andInvestment Behaviour of Small Farmers," Indian Journal of AgriculturalEconomics, Vol. XXVII, October-December 1972, pp. 43-50.

8. Datta Roy Choudhury, Uma: "Income, Consumption and Saving in Urbanand Rural India," The Review of Income and Wealth, Vol. 24, March 1968.

9. Datta Roy Choudhury, Uma: "Industrial Breakdown of Capital Stock inIndia," The Journal of Income and Wealth, Vol. 1, No. 2, April 1977.

10. Directorate of Economics and Statistics, Ministry of Agriculture andIrrigation: Seventh Livestock Census, New Delhi, 1951.

11. Directorate of Economics and Statistics, Ministry of Agriculture andIrrigation: Eighth Livestock Census, New Delhi, 1956.

12. Directorate of Economics and Statistics, Ministry of Agriculture andIrrigation: Ninth Livestock Census, New Delhi, 1961.

13. Directorate of Economics and Statistics, Ministry of Agriculture andIrrigation: Tenth Livestock Census, New Delhi, 1966.

14. Directorate of Economics and Statistics, Ministry of Agriculture andIrrigation: Eleventh Livestock Census, New Delhi, 1972.

15. Divatia, V.V.: "Inequalities in Asset Distribution of Rural Households,"Reserve Bank Staff Occassional Papers, Vol. 1, Issue No. 1, June 1976.

Page 45: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 42 -

16. FAI 1978: Fertilizer Association of India, "Annual Review of FertiliserConsumption and Production 1977-78," New Delhi, July 1978.

17. Gupta, K. L.: "On Some Determinants of Rural and Urban Household SavingBehaviour," Economic Record, Vol. 46, December 1970.

18. Kahlen, A. S., H.S. Bal and G. Singh: "Savings and Investment Patternsin Farm Families in the Punjab (India)," Indian Journal of AgriculturalEconomics, Vol. XXVII, October-December 1972, pp. 19-29.

19. Lal, Ram N.: Capital Formation and its Financing in India, AlliedPublishers, Bombay, 1977.

20. Lewis, W. Arthur: "Economic Development with Unlimited Supplies ofLabour," The Manchester School of Economic & Social Studies, Vol. XXII,1956.

21. Misra, B. and S. C. Mallick, "Factors Influencing Capital Formation inAgriculture," Indian Journal of Agricultural Economics, Vol. XXIV,October-December 1969, pp. 93-105.

22. Mukherjee, M. and N. S. R. Sastry: "An Estimate of the Tangible Wealthof India" in Goldsmith, R and C. Saunders (ed.): The Measurement ofNational Income, Income and Wealth Series VIII, Bowes & Bowes, London,1959.

23. National Council of Applied Economic Research: All India Rural House-hold Survey 1962, Vol. II, New Delhi, 1965.

24. National Council of Applied Economic Research: All India Rural House-hold Survey of Income, Saving and Consumer Expenditure (With SpecialReference to Middle Class Households), 1967-68, New Delhi, 1972.

25. National Council of Applied Economic Research: Changes in Rural Incomein India 1968-69, 1969-70, 1970-71, New Delhi, 1975.

26. Rajagopalan V. and S. Krishnamoorthy, "Savings Elasticities andStrategies for Capital Formation - A Micro Analysis," Indian Journal ofAgricultural Economics, Vol. XXIV, October-December 1969, pp. 110-116.

27. Reserve Bank of India: "Estimates of Tangible Wealth in India,"R.B.I. Bulletin, January 1963.

28. Reserve Bank of India: "Estimates of Saving and Investment in the IndianEconomy: 1950-51 to 1962-63," R.B.I. Bulletin, March 1965.

29. Reserve Bank of India: "Estimates of Tangible Wealth in India,"R.B.I. Bulletin, October 1972.

30. Reserve Bank of India: All India Debt and Investment Survey 1971-72,Vol I, Bombay, April 1975.

Page 46: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 43 -

31. Shah, S. L. and L. R. Singh, "Capital Formation in Agriculture of theTerai Region of Uttar Pradesh," Indian Journal of Agricultural Economics,Vol. XXIV, October-December 1969, pp. 87-92.

32. Singh 1978: Inderjit Singh, "Small Farmers and the Landless in SouthAsia: Prospects and Problems," World Bank, Washington, August 1978.

Page 47: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

RECENT PAPERS IN THIS SERIES

No. TITLE OF PAPER AUTHOR

350 Energy Options and Policy Issues in D. Fallen-BaileyDeveloping Countries T. Byer

351 Growth and Equity in Semi-Industrialized J. BergsmanCountries

352 Capital Flows and Developing Country Debt J. Katz

353 Trade Policy for Developing Countries D. Keesing

354 Development Problems of Mineral-Exporting G. NankaniCountries

355 The Global Framework R. Cheetham, S. GuptaA. Schwartz

356 The Distribution of Income in Brazil G. Pfefferman, R. Webb

357 Estimating Shadow Prices for Colombia in an W. SchohlInput-Output Table Framework

358 Inter-Country Comparison of "Real" (PPP) P. IsenmanIncomes: Revised Estimates and UnresolvedQuestions

359 Price Distortions in Agriculture and Their E. LutzEffects: An International Comparison M. Bale

360 Costs and Benefits of Agricultural Research: G.E. SchuhThe State of the Arts H. Tollini (consultants)

361 Investment in International Agricultural G. ScobieResearch: Some Economic Dimensions (consultant)

362 Identification and Appraisal of Rural Roads H. BeenhakkerProjects A. Chammari

363 Small Enterprises in African Development: J. Page (consultant)A Survey

364 Income, Consumption and Poverty in Thailand, 0. Meesook1962/63 to 1975/76

365 A Survey and Critique of World Bank Supported R. MarrisResearch on International Comparisons of Real (consultant)Product

Page 48: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

- 2-

No. TITLE OF PAPER AUTHOR

366 Paradigms in the Study of Urban Labor Markets D. Mazumdarin LDCs: A Reassessment in the Light of anEmpirical Survey in Bombay City

367 Incentives for Resource Allocation: S. AcharyaA Case Study of Sudan

368 Why the Emperor's New Clothes Are Not Made D. Morawetzin Colombia (consultant)

369 Economic and Social Analysis of Projects K. Cleaverand of Price Policy: The Morocco FourthAgricultural Credit Project

370 The Tokyo Round and the Developing Countries B. Balassa

371 Bus Ownership and Efficiency in Urban Areas A. WaltersC. Feibel

372 The Tokyo Round: Results and Implications R. Kemperfor Developing Countries (consultant)

373 Malnutrition: Some Measurement and Policy Issues T.N. Srinivasan

374 The Prevalence of Calorie Deficient Diets S. Reutlingerin Developing Countries H. Alderman (consultant)

375 The Design of Organizations for Rural F. Lethem, B. ThoolenDevelopment Projects - A Progress Report W. Smith (consultant)

376 Fostering the Capital-Goods Sector in LDCs: H. Pack (consultant)A Survey of Evidence and Requirements

377 Macroeconomic Implications of Factor H. Pack (consultant)Substitution in Industrial Processes

378 Household Income or Household Income per Capita G. Dattain Welfare Comparisons J. Meerman

379 Managing Information for Rural Development: G. DeboeckLessons from Eastern Africa B. Kinsey

380 Women in the Urban Labor Markets of Africa: N. ShieldsThe Case of Tanzania

381 Some Aspects of Wheat and Rice Price Policy R. Krishnain India D.S. Raychaudhuri

(consultants)

Page 49: Trends in Rural Savings and Private Capital Formation in India · composition of phiysical capital. Trends in this composition are also dis-cussed in Section 6. 2. Growth of Saving

HG3881.5 .W57 W67 no.382 c.3

Krishna, Raj.Trends in rural savings and

private capital formation in

India /

t~.~fATEj NAME AND EXTENSION ROOMNUMBER