Indebtedness of Cultivator Households
Some Puzzling Results
Two important surveys, viz, the Situation Assessment Survey and the All-India Debt and Investment Survey were conducted in the 59th round of the National Sample Survey Organisation in 2003. A common topic was the indebtedness of farmer/cultivator households. The estimates of incidence of debt, extent of indebtedness and also the pattern of debt owed to institutional and non-institutional agencies, showed wide variations among different states and also at the all-India level in the two surveys, mainly because of the differences in concepts, definitions, methods of data collection, sample design, etc.
K G K SUBBA RAO
T
Economic and Political Weekly August 12, 2006
some of the important issues, which resulted in conflicting versions on indebtedness.
Scope and Coverage of the Surveys
As a prelude to the discussion, it would be appropriate to list out the differences in the two surveys under reference. As per SAS 2003, a farmer household is defined as one possessing some land and engaged in agricultural activities during the last one year, whereas in AIDIS 2002-03 all households operating at least 0.002 hectares of land during 2002-03 were treated as cultivator households, while those operating no land or less than this cut-off point were considered as non-cultivator households. Thus it would appear that some of the households operating land less than the cut-off point of 0.002 hectares classified as non-cultivator households in AIDIS 2002-03, could have been classified as cultivator households in the SAS 2003. Thus there could be an overlap in the classification of the households in the lowest strata, but this per se, will not vitiate comparison of the estimates and the structure of debt in the two surveys.4
Though the survey period was the year 2003 for both the surveys, the reference period and the method of collection of data were different for each of the surveys. Indebtedness for SAS 2003 was recorded as on the date of survey. Thus it was a moving reference date for each of the households canvassed on different dates, and as such, the data cannot be stated to be outstanding as on a particular reference date. In contrast, the data on assets and liabilities of the households for AIDIS 2002-03 were collected in two visits. The first visit to the sample households was made during the first eight months of the survey period (January-August), while the second visit in the next four months. The estimates of cash loans outstanding as on June 30, 2002, were based on the first visit data by recording outstanding loans as on date of survey and working backwards, by subtracting borrowings and adding repayments during the intervening period. Thus the reference period for collection of data on loans was brought to a common platform by eliciting the requisite details from the households on borrowings and repayments.
In the erstwhile decennial surveys of the AIDIS up to 1981-82, indebtedness of the households comprised not only cash loans, but also current liabilities outstanding as on date of survey, i e, a household was considered as indebted if some amount of cash loan and/or current liabilities5 were outstanding as on the date of survey. All these data were collected in a single block under the caption “cash dues payable”. However, since AIDIS 1991-92, information on current liabilities and data on cash loans were collected in different blocks of the schedule. It may be noted that current liabilities recorded as on date of survey cannot be worked back to the reference date as on June 30 as in the case of outstanding loans, which could be brought to uniform reference date, as indicated above. The rationale in bifurcating the cash loans and current liabilities was based on the premise that the various characteristics such as rate of interest, duration of loan, credit agency, purpose, etc, were more relevant for loans; these may not be relevant for current liabilities, which were dues payable by the households in a short period of time and interest rate is generally not invoked.6 Thus there was a departure in the method of data collection of the two constituent items, which were hitherto considered as a single entity. In the AIDIS 2002-03 also, the same procedure was adopted. Strictly speaking, current liabilities recorded as on date of survey and the outstanding loans as on June 30 are not additive, as they relate to different time points. Even if the duration of the current liabilities and the source of agency were recorded in the schedule, these data cannot be mingled with loans, as working out these liabilities backwards as in the case of loans, is fraught with difficulties, distorting the results to some extent. Though the departure in methodology was an improvement, comparison of the data with that of the previous benchmark years of the decennial surveys was not feasible.7 Viewed in the above perspective, it is desirable to record the two constituents of indebtedness separately.
In contrast to the methodological details followed in the AIDIS surveys cited above, the SAS 2003, which was also conducted in the same 59th round, adopted the erstwhile definition of cash debt. A farmer household for this survey was defined as indebted, if the liability of the household was greater than or equal to Rs 300, which implicitly includes current liabilities and/ or loans. Thus the different definitions on indebtedness adopted in the two surveys give rise to controversial estimates on
Table 1: Proportion of Cultivator Households Reporting Cash Loans and Current Liabilities and Average Amount (Rs) Per Household (A) as at the End of June 1991 and 2002
Cash Loans Current Liabilities Total Debt P APAP A
1991 25.9 2294 13.2 252 34.6 2546 2002 29.7 9261 12.0 562 @ 9823
Notes: P: Proportion of cultivator households; A: Current liabilities and average amount per household; @: Not available.
Sources: (1) ‘AIDIS,1991-92 –Incidence of Indebtedness of Households’,Reserve Bank of India Bulletin, February 2000; (2) NSSO Report No 501: Household indebtedness in India as on June 30, 2002.
Table 2: Indebtedness of Cultivator Households
(Proportions in per cent and amount in rupees)
| Farmer Households | Cultivator Households | AIDIS 2002-03 | |
|---|---|---|---|
| (SAS 2003) | (June 2002) | ||
| State | P R APH | APRH P R | APH APRH |
| Andhra Pradesh | 82.0 23965 | 29226 54.0 | 16154 29915 |
| Assam | 18.1 813 | 4492 6.7 | 641 9567 |
| Bihar | 33.0 4476 | 13564 22.5 | 3336 14827 |
| Gujarat Haryana Himachal Pradesh | 51.9 15526 53.1 26007 33.4 9618 | 29915 33.9 48977 31.7 28796 17.9 | 12958 38225 17340 54700 5843 32643 |
| Jammu and Kashmir | 31.8 1903 | 5984 3.8 | 1198 31526 |
| Karnataka | 61.6 18135 | 29440 39.1 | 13422 34327 |
| Kerala | 64.4 33907 | 52651 42.9 | 27641 64431 |
| Madhya PradeshMaharashtra | 50.8 14218 54.8 16973 | 27988 31.7 30863 37.8 | 12246 38631 14268 37746 |
| Orissa | 47.8 5871 | 12282 31.3 | 3976 12703 |
| Punjab RajasthanTamil Nadu | 65.4 41576 52.4 18372 74.5 23963 | 63572 28.5 35061 36.7 32165 40.3 | 25211 88459 13261 36134 14823 36782 |
| Uttar Pradesh | 40.3 7425 | 18424 24.1 | 5363 22253 |
| West Bengal All India | 50.1 5237 48.6 12585 | 10453 24.7 25895 29.7 | 3820 15466 9261 31182 |
Notes: PR: Proportion of households reporting; APH: Average debt/loan per household; APRH: Average debt/loan per reporting household (derived) = APH/PR.
Source: NSSO Reports No 498 and 501.
Economic and Political Weekly August 12, 2006 incidence of debt (proportion of households reporting indebtedness) and the extent of indebtedness.
It would be relevant at this juncture to assess the impact of current liabilities in the total outstanding cash debt in the last two AIDIS surveys, where separate details were available. Table 1 provides the details in this regard.
It may be observed from the above data that the share of current liabilities in total debt was 6 per cent for the cultivator households at the end of June 2002, as against 9.9 per cent at the end of June 1991.The structure of debt will undergo change if current liabilities are taken into account. Taking the specific example of cash debt according to credit agency, it would appear, on a priori grounds that the share of debt owed to institutional agencies will be deflated, with concurrent increase in non-institutional sources, as current liabilities pertain to this category. The proportion of households reporting non-institutional sources will also vary depending upon the households reporting current liabilities only. The tabulations generated by the RBI for the year 1991 indicate that the share of debt owed to institutional agencies is tilted down for each of the institutional agencies, when cash debt covers both loans and current liabilities. These aspects are discussed subsequently.
Findings of the Two Surveys
Against the above backdrop, we may analyse the differences in the estimates of incidence of indebtedness (reflected by the proportion of households reporting indebtedness ( PR) and the extent of indebtedness (as reflected in the average amount of debt (APH) and average per reporting household (APRH). The APRH is a better indicator than the APH, for a comparative picture of indebtedness in the two surveys under reference. Table 2 provides the statewise details in this regard. It may be observed that about 49 per cent of the farmer households were indebted as per SAS 2003, as against
29.7 per cent at the end of June 2002, as per the AIDIS 2002-03. This was so in respect of all the states, with more than 50 per cent of the farmer households reporting debt in the SAS; this was almost half of this in the AIDIS 2002-03. The differences in the incidence of debt were observed to be higher, particularly in states, viz, Andhra Pradesh, Jammu and Kashmir, Tamil Nadu, the AIDIS survey showing considerably lower proportions of indebted households. The average debt per reporting household for the country as a whole was higher in the AIDIS at Rs 31,182, compared to Rs 25,895 in the case of farmer households in the SAS. These estimates dimensionally tally in respect of Andhra Pradesh, Bihar, Maharashtra, Orissa and Rajasthan, while major differences were noticed in respect of the other states.
The share of different credit agencies in total debt/loans, reported in the two surveys is presented in Table 3.
It may be observed that the share of debt owed to institutional agencies was slightly higher at 61.1 per cent in the case of AIDIS 2002-03, as against 57.7 per cent in SAS 2003. However, the share of debt owed to commercial banks in respect of SAS 2003 was higher (35.6 per cent) than that of cooperative banks (19.6 per cent), while a reverse pattern was noticed for cultivator
INDIAN INSTITUTE OF CAPITAL MARKETS
(Formerly UTI Institute of Capital Markets)
announces
TENTH CAPITAL MARKETS CONFERENCE (December 18-19, 2006)
The 10th Capital Markets Conference (CMC) will be held at Indian Institute of Capital Markets, Vashi during December 18-19, 2006.
Conference Theme
The 10th Capital Markets Conference invites research papers in the following six conference tracks:
Research papers and participation as delegates are invited from
Paper Submission Procedure
Dates to remember
For further queries contact:
Ms. Jaisheela Sane, Conference Secretariat, Indian Institute of Capital Markets, UTI House, Post Box. 99, Plot No - 82, Sector - 17, Vashi, Navi Mumbai - 400705. India. Telephone No: 022-27883000(B)/27883011(D) Fax No: 022-27896863/2824 Email: cmc@utiicm.com Website: www.utiicm.com
Economic and Political Weekly August 12, 2006
households for AIDIS 2002-03 (30.2 per cent in respect of cooperative banks and
26.3 per cent in respect of commercial banks. With the commercial banks playing a predominant role in financing agricultural credit and the targets laid down for the subsector of agriculture in priority sectors, the structure of debt of farmer households shown against commercial banks in SAS 2003 seems to be more credible. The divergence in the pattern of debt shown against commercial and cooperative banks in the two surveys needs further investigation. It is interesting to note from both the surveys that the share of moneylenders in the non-institutional sources accounted for more than one-fourth of the total debt, which was substantially higher than that observed in AIDIS 1991-92 (17.5 per cent).8 The available data on outstanding institutional credit to agriculture confirms the predominant role of commercial banks in purveying credit to the cultivator households.
Thus it may be observed from Table 4 that the share of commercial banks (including regional rural banks ) has been showing a consistently upward trend over the decades. Though the above data are not separately available for rural and urban population groups, it could be assumed that these are, by and large, to the rural cultivator households, though a small part
Table 3: Share of Different Credit Agencies in Total Debt/Loans in SAS 2003 and AIDIS 2002-03
Credit agency SAS 2003 AIDIS (June 2002)
Government 2.5 1.7 Cooperative banks
and societies 19.6 30.2 Commercial banks 35.6 26.3 Other institutional agencies – 2.9 All institutional agencies 57.7 61.1 Landlord – 0.9 Agricultural moneylender 25.7@ 9.9 Professional moneylender 16.9 Relatives and friends 8.5 6.2 Others* 8.2 5.0 All non-institutional agencies 42.4 38.9 All agencies 100.0 100.0
Notes: *Others includes doctors, traders, etc. @ Covers both agricultural and professional moneylenders. Break-up is not available.Source: NSSO Reports 498 and 501
of the credit may be going to the noncultivator households.
Concluding Observations
The study examined the differences in the estimates of the incidence of debt and the extent of indebtedness, emanating from the two surveys, viz, SAS 2003, and AIDIS 2002-03, which were conducted in the same year under the subject programme of the 59th round of the NSSO. The differences were observed to be mainly due to the concepts, definitions and the methods of data collection in the two surveys. To summarise, about 49 per cent of the farmer households were reported to be indebted in the SAS, as against 30 per cent in the case of the AIDIS. The incidence of debt was more than 50 per cent in most of the states for SAS, as against less than 25 per cent for the AIDIS. The average debt per reporting household for the country as a whole was higher in the AIDIS, at Rs 31,182, compared to Rs 25,895 in the SAS. Wide interstate differences were also noticed in these estimates in both these surveys. The share of debt of cultivator households owed to institutional agencies was broadly the same in both the surveys, though it was slightly higher at 61 per cent in the case of AIDIS. As between commercial and cooperative banks, the share of debt owed to commercial banks was higher at 35.6 per cent than that of the cooperative banks (19.6 per cent) in the SAS. In contrast, the AIDIS results showed a higher share for cooperative banks (30.2 per cent ), than for commercial banks (26.3 per cent). The available evidence from the supply side lends support to the SAS results, with commercial banks accounting for more than two-thirds of the total institutional credit for agriculture and allied activities.
Thus a comparative picture of the indebtedness of the farmer/cultivator households emanating from the two surveys of the 59th round of the NSSO showed wide variations in the estimates. As indebtedness is an area of importance for formulating schemes of financial assistance to the weaker sections in the agricultural sector,
Table 4: Direct Institutional Credit for Agriculture and Allied Activities(Total Short-term and Long-term Loans Outstanding)
(Rs crore)
| Cooperatives | Commercial Banks | Regional Rural Banks | Total | |
|---|---|---|---|---|
| 1981-82 1991-92 2001-02 | 4821.0 (55.8) 12176.4 (39.1) 25321.0 (32.2) | 3541.1 (41.0) 16981.1 (54.5) 45105.5 (57.3) | 273.1 (3.2) 1984.3 (6.4) 8286.0 (10.5) | 8635.2 (100.0) 31141.8 (100.0) 78712.5 (100.0) |
Note: Figures in the brackets are the per cent shares in the total. Source: Reserve Bank of India, Handbook of Statistics on the Indian Economy, 2004-05.
it is absolutely necessary to adopt uniform concepts and definitions. As indebtedness of the cultivator households is covered in the AIDIS surveys even in the earlier rounds, it may be advantageous to dovetail such requirements from other organisations in decennial surveys on debt and investment. This is desirable from the consideration of optimum utilisation of resources and cost considerations, inevitable in largescale sample surveys, and also to avoid controversy on the estimates generated.
EPWEmail: kgksubbarao@gmail.com
Notes
1 The NSSO brought out the following reports on the SAS 2003: Consumption expenditure of farmer households (Report No 495); Some aspects of farming, 2003 (Report No 496); Income, expenditure of farmer households (Report No 497); Indebtedness of farmer households (Report No 498); Access to modern technology for farming, 2003 (Report No 499).
2 The following Reports were brought out by the NSSO on the AIDIS 2002-03: household assets and liabilities in India (Report No 500); Household indebtedness in India as on June 30, 2002 (Report No 501); Household borrowings and repayments in India (Report No 502); Household asset holdings and indebtedness by social groups (Report No 503); Household capital expenditure in India (Report No 504).
3 The reference period for the survey was 200203 and as such, it cannot be called a decennial survey as in the case of the erstwhile surveys of the AIDIS 1991-92 et al, which had an interval of 10 years between the surveys.
4 The distribution of farmer households according to size of land possessed from the SAS 2003 indicated that the farmer households possessing land less than 0.01 hectares accounted for only
1.4 per cent of the total estimated number of farmer households. This aspect was discussed in (Note 5) also.
5 Loans taken from various agencies (institutional and non-institutional) covered all loans at a specific rate of interest for a specific period of time. However, if a loan is contracted without rate of interest, it was considered as a loan from friends and relatives. Instances of loans taken at zero rate of interest were also not uncommon in agrarian credit markets, where there is a linkage between land, labour and credit markets. As distinguished from loans, current liabilities are of a short-term duration and recurring from time to time.These include dues payable, amounts due to provision merchants, trade credit, unpaid bills of doctors, lawyers, etc.
6 In the case of trade credit, however, there is an implicit rate of interest depending upon the terms of trade between the buyer and seller of goods and the duration of the trade credit.
7 In order to facilitate comparison with the earlier AIDIS survey data , the Reserve Bank of India (RBI ) generated additional tabulations on cash debt (including current liabilities) for the AIDIS 1991-92 , based on the primary data of individual sample households supplied by the NSSO. For further details, reference may be made to Note 4.
8 See NSSO Report No 420 (1) for details.
Economic and Political Weekly August 12, 2006
