Magnitude and Determinants of Reverse Tenancy: A Study of Agriculturally Developed and Backward States

Abstract

The present study analyses the extent of reverse tenancy in terms of who leases in and who leases out land in agriculturally developed and backward states using unit-level National Sample Survey data from different rounds. There are three important findings of the study. First, reverse tenancy in terms of distribution of leased in and leased-out land among households of different size categories is more pronounced in agriculturally developed states while in agriculturally backward states, most of the leased in and leased-out land was accounted for by sub-marginal, marginal and small households. Second, in terms of concentration of leased in and leased-out land among households at different levels of ownership hierarchy, the tenancy relations in both the categories of states conform to the traditional variety where most of the leased-in land is concentrated among households that are at the bottom and middle levels and leased-out land among households that are at top different levels. Third, the results of the logit regression further show that variables such as household size, age of the head of family, education of the head of the family, self-employment in agriculture, ownership of livestock and land use affect the probability of medium and large households leasing in and sub-marginal, marginal and small households leasing out land though the nature of their effect, and statistical significance of the regression coefficients vary among agriculturally developed and backward states and also over the years.

Keywords

Reverse tenancy agriculturally developed backward states leased-in leased-out

Introduction

Indian economy continues to be agrarian notwithstanding numerous changes on technological, institutional and demographic fronts during the last 60–70 years. For example, even today, more than three-fifths of the total rural workforce is engaged in agriculture for livelihood, and land as an asset accounts for more than 70% of the total rural assets. In such economies, ownership of land is synonymous with power, social status, privileges, control and access to local-level institutions and so on, and its distribution largely determines income distribution. The extremely skewed distribution of land in terms of a preponderant majority of rural households owning extremely small proportion of the total land gives rise to various kinds of rural institutions, most notably, tenancy, mode of wage payment, emergence of agricultural labour as a class and so on. In such economies, in the absence of alternative employment opportunities, rural households, in particular landless, sub-marginal and marginal households, lease in land from medium and large households under different terms of tenancy to earn their livelihood. These tenancy arrangements are largely characterised by insecure tenancies, exorbitant rents and share cropping with interlinkages in land labour and credit markets having far-reaching implications for agricultural development on the one hand and wellbeing of the tenants on the other. The tenancy relations, however, change over time in response to institutional, technological and demographic changes. Some of these changes are rapid development of secondary and tertiary sectors coupled with increasing modernisation and commercialisation of agriculture. As a consequence, rural households who have extremely small and tiny holdings tend to lease out their land and switch over to wage paid jobs that become increasingly available in the rural and urban areas, giving rise to a phenomenon of ‘reverse tenancy’. Further, while traditional pattern of tenancy is compulsive and oblige small and marginal farmers to lease in land for sustaining their livelihoods in the absence of alternative job opportunities on onerous and exploitative terms of tenancy, reverse tenancy which implies leasing out land by these households to medium and large farmers is a voluntary decision of such farmers to switch over to non-farm employment as and when it becomes available. This type of tenancy is considered to be promoting both productive and allocative efficiencies and augur well for agricultural development.

The extent of reverse tenancy in Indian agriculture has been a point of discussion in the literature since the early 1970s when the prevalence of the practice of medium and large farmers leasing in land and marginal and small households leasing out land was reported. The studies reporting phenomenon of reverse tenancy found that with the increasing spread of new agricultural technology and rising commercialisation of agriculture, medium and large households enter lease market as lessees to make optimal use of their capital inputs such as tractors, threshers, pump sets and so on and reap economies of scale to maximise profits. On the other hand, marginal and small households have their own compulsions to opt out of cultivation because of, inter alia, small amount of owned land, assured employment on farms of medium and large farmers and availability of non-farm employment opportunities and find it more rewarding to lease out their land and switch over to wage paid manual employment (Bharadwaj & Das, 1975; Dasgupta, 1984; Jodha, 1981; Nadkarni, 1976; Rao, 1974; Ray, 1978; Shergill, 1989; Singh, 1989; Vyas, 1970). The factors which were reported to have been forcing out marginal and small households out of cultivation in the 1970s and the 1980s have further become more important because of numerous changes in the rural economy in general and in agriculture in particular since the early 1990s. These changes are widespread adoption of new agricultural technology across crops and regions, ongoing demand-driven diversification in cropping pattern and commercialisation of agriculture, sub-division of holdings, migration of rural households to urban areas as a consequence of increase in rural non-farm employment opportunities, thereby lessening the burden on agriculture sector, increase in the disparities between wage earnings in the agricultural and non-agricultural sectors and also increase in the ratio of income per non-agricultural worker to farm income per cultivator (Chand et al., 2017; Papola, 2014). Again, in more recent times, Indian agriculture is also witnessing increasing uncertainty due to erratic weather conditions coupled with depleting ground water level and growing agrarian distress manifested in rising cost of production, falling output prices, falling incomes, increasing indebtedness among farmers, and ultimately, increasing farmers suicides (Sharma & Malik, 2019). These changes, combined with an extremely unequal distribution of land, are forcing small and marginal households to seek wage paid employment in farm and non-farm sectors. According to situation assessment survey 2013, 40% of the farmers want to quit agriculture on the first available opportunity in other sectors of the economy. In fact, some recent studies have reported that households of lower farm size categories such as sub-marginal, marginal and small are increasingly opting to lease out their land to medium and large households because, among other things, cultivation of crops on extremely small holdings is increasingly becoming costly and non-remunerative. These studies have also reported that these patterns are more pronounced in agriculturally developed regions as compared to agriculturally backward regions ( Bellemare, 2007; Iyer, 2009; Murty, 2004; Ray, 2016; Sharma, 2000, 2006; Sharma, 2005; Singh & Sharma, 2013; Swain, 1999; Vijay, 2012; Vijay & Sreenivasulu, 2013).

It is against this background that the present study analyses the extent of reverse tenancy and the factors determining leasing in of land by medium and large households and leasing out land by sub-marginal, marginal and small households in agriculturally developed and backward states of India using National Sample Survey (NSS) unit-level data from different rounds. The article is divided into four sections. The first section discusses the data and methods used in the study. The second section discusses the distribution of ownership holdings among households of different size categories followed by the extent of reverse tenancy in both the categories of states. The results of logit regression quantifying the effect of different factors on the probability of sub-marginal, marginal and small households leasing out land and medium and large households leasing in land are presented and discussed in the third section. Last section gives summary and conclusions.

Data and Methods

The study is based on NSS unit-level data for three decennial rounds, that is, 48th (1991–1992), 59th (2002–2003) and 70th (2012–2013) on Land and Livestock Holdings. The unit-level data that was accessed from the Ministry of Statistics and Programme Implementation, New Delhi, which was available in CDs in .txt format. The datasets were generated by retrieving data with the help of Stata software package individually for all three rounds. Also, the data available in different NSS Reports on Ownership and Operational holdings were used.

In order to select the two states representing different levels of agricultural development, namely, agriculturally developed and the backward states, the data on the average yield of food grain from 2009–2010 to 2013–2014 for 28 states were collected from the RBI Handbook of Statistics on Indian Economy, 2015–16. The three different series of averages, that is, M1, M2 and M3 on average yield of all states from 2009–2010 to 2013–2014 were created by using the moving average method in MS Excel (Malik, 2019). All the three series were arranged in descending order from the highest to the lowest levels of productivity. It was found that among all the states Punjab occupied the first position followed by Haryana and West Bengal in top 14 states. Similarly, six major states, namely, Jammu and Kashmir, Karnataka, Odisha, Rajasthan, Maharashtra and Madhya Pradesh came among the bottom 14 states. Since a number of studies on different aspects of Punjab agriculture are available, we preferred to choose Haryana from the top 10 states to represent agriculturally developed states. Among the bottom 14 states, while tenancy is completely banned in J&K, it is prohibited in Himachal Pradesh with some notable exceptions. Therefore, Odisha was chosen to represent agriculturally backward states.

Further, it may be mentioned here that in the NSS reports on landholdings, the data on the tenancy is available for households of different size categories classified both on the basis of their ownership holdings and operational holdings. Therefore, the changes in the incidence of reverse tenancy among households of different categories can be analysed in terms of both ownership holdings and operational holdings. However, the classification of households into different size classes on the basis of operational holdings may not reveal their actual land ownership status as it consists of land operated by them which may include land leased in and exclude the land leased out. It may be mentioned here that from the policy perspective the classification of households on the basis of their land ownership status is more important inasmuch as the benefits of some schemes are available to households owning less than a certain amount of land. For example, the benefits of Pradhan Mantri Kisan Samman Nidhi (PM-KISAN) are available to those households who own less than 2 hectare of land while tenant households are excluded from the scheme. Again, the classification of households on basis of operational holdings does not provide any information on the amount of land leased in by the landless households, that is, those households who do not own any land or own less than 0.002 hectare. Therefore, to see the extent of leased-in land among households of lower size categories such as landless and sub-marginal households and how it has changed overtime, we have used classification of households into different size categories based on the ownership holdings.

Further, to estimate changes in the concentration of leased-in and leased-out land at different levels of land ownership hierarchy in two states, Lagrange interpolation polynomial method was used to compute the share of those at the top 1%, 5%, 10% and 20%, 30% at the middle level and 50% at the bottom in the FORTRAN.¹

The Lagrange’s form of interpolating polynomial is given below (Carnham et al., 1969, pp. 27–29).

P_{n} (x) = \sum_{i = 0}^{n} L_{i} (x) f (x_{i})

\begin{array}{l} \bar{y} (\bar{x}) = \sum_{i = m i n}^{m i n + d} L_{i} (\bar{x}) y_{i} \\ \begin{array}{l} Where L_{i} (\bar{x}) = Π_{\underset{j \neq 1}{j = 0}}^{n} \frac{\bar{X} - X_{j}}{X_{i} - X_{j}}, i = 0, 1, 2, \dots \dots \dots, n \end{array} \end{array}

i = min, min + 1, ……, min + d.

Two logit regression models have been estimated for each of the two states to identify factors that determine the probability of medium and large households leasing in and small and marginal households leasing out land. In the logistic regression model, the dependent variable is binary or dichotomous and contains data coded as 1 or 0. The functional form of logistic regression equation is given below:

logit (p) = α_{0} + α_{1} x_{1} + α_{2} x_{2} + \dots \dots + β_{k} x_{k}

where x₁, x₂, x_k are explanatory variables.

Distribution of Ownership Holdings

Table 1 shows the changing distribution of ownership holdings in Haryana, an agriculturally developed state, and Orissa, an agriculturally backward state. As may be seen from the table, distribution of landholdings in Haryana, measured by the Gini ratio, became less skewed during the period as was evident from the decrease in the value of Gini ratio from 0.59 in 1992 to 0.55 in 2013. In terms of percentage changes in the distribution of holdings and area owned among different categories of households, the table shows that while there has been a decrease in the proportion of land accounted for by sub-marginal and marginal households over time, the percentage of land owned by the small and medium households has increased. More importantly, however, between 1992 and 2013, there was a significant decrease in the percentage share of area owned by large households owning more than 4 hectare of land. In comparison, the distribution of landholdings in Orissa became more unequal as the value of Gini ratio increased from 0.47 in 1992 to 0.52 in 2013. In terms of changes in the percentage share of land owned by different categories of households, the table shows that decrease in the percentage share of land owned by sub-marginal and marginal households was accompanied by a huge increase in the proportion of land owned accounted for by small and medium households. Similarly, the percentage of land owned by large households recorded a significant decrease over the period. Thus, the distribution of land in agriculturally backward state changed in favour of small and medium households over a period of time. The extremely unequal distribution of land in both the categories of state with a preponderant majority of households owning extremely low proportion of total land in comparison to their numerical proportions, coupled with above-mentioned recent changes in the Indian economy in general and in agricultural sector in particular, has implications for leasing in and leasing out land by different categories of households. It is against this background that it would be interesting to know as who leases in and who leases out land in agriculturally developed and backward state. This we discuss in the following sections.

Table 1.

Distribution of Households and Area Owned Among Different Categories of Households in Agriculturally Developed and Backward States, 1992–2013 (%).

Category of households	Haryana						Odisha
	1992		2003		2013		1992		2003		2013
	H	A	H	A	H	A	H	A	H	A	H	A
Landless	3.69	0.00	9.94	0.00	2.98	0.00	13.82	0.00	7.37	0.00	5.33	0.00
Sub-marginal	44.85	2.51	60.68	5.05	65.62	4.58	45.91	23.88	66.18	20.43	50.76	8.61
Marginal	10.48	5.43	8.13	7.90	7.35	7.35	15.97	20.63	13.00	21.64	9.52	9.06
Small	13.49	13.40	9.26	16.30	10.06	19.59	14.42	24.20	9.20	27.46	24.25	42.08
Medium	18.20	33.54	7.04	24.33	10.57	37.06	7.35	18.85	3.38	19.37	8.61	29.66
Large	9.29	45.11	4.97	46.42	3.42	27.16	2.53	12.44	0.86	11.10	1.53	10.58
All households	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0
Gini ratio	0.59		0.59		0.55		0.47		0.49		0.52

Source: NSSO unit-level data, 48th, 59th and 70th rounds of Land & Livestock Holdings.

Note: Landless who own less than 0.002 ha; sub-marginal between 0.002–0.5 ha; marginal 0.5–1 ha; small 1–2 ha; medium 2–4 ha and large above 4 ha.

H = Households, A = Area

Extent of Reverse Tenancy

The extent of reverse tenancy, measured in terms of distribution of land leased in and leased out among households of different farm size categories, in agriculturally developed and backward states during 1992–2013 has been shown in Table 2. The table shows that in agriculturally developed state of Haryana, share of leased-in land accounted for by the households of lower farm size categories (below 2.00 hectare) was 46.24% in 1992 which increased to 66.29% during 2003 but further declined hugely to 34.81% in 2013. However, the percentage share of area leased in accounted by medium and large households registered a large increase from 53.76% in 1992 to nearly 65.19% in 2013. Among these households, large households, owning more than 4 hectare of land, alone accounted for 50% of total leased-in land which was around one-third in the early 1990s. The change in the distribution of leased-in land over the years among different categories of households is, however, half the story. To know whether reverse tenancy exists or not, we also need to examine changes in the percentage distribution of leased-out land among households of different size categories. And, in this regard, the table shows that there was a huge increase in the percentage share of lower categories households, namely, sub-marginal, marginal and small households in the total leased-out land which increased to around 47% in 2013 from around 7% in 1992. It is also important to mention that more than half of the leased-out land (around 52%) in 2013 was still contributed by medium and large households. It, however, needs to be underlined that since households in lower size category account for a low proportion of the total owned land, their share in the total leased-out land is expected to be low as compared to their large counterparts. These findings, therefore, suggest that the phenomenon of reverse tenancy is more pronounced in agriculturally developed state. In agriculturally backward state of Odisha, the data shows that there has not been much change in the distribution of leased-in land among households of different size categories and that in 2013 nearly 90% of the leased-in land was accounted for by landless, sub-marginal and marginal households owning less than 1 hectare of land. Insofar as the distribution of leased-out land was concerned, it was interesting to see that again around 70% of the total leased-out land was also contributed by households of these categories. The data thus suggests that, in agriculturally backward states, the business of leasing in and leasing out land is largely confined to households of lower size categories contrary to the prevailing belief that in such states while lessors are primarily medium and large households, lessees belong to lower farm size categories. In brief, the tenancy relations in an agriculturally backward state in terms of distribution of land leased in and leased out conform neither to the traditional type where most of the area leased in is accounted for by households of lower size categories and leased out by medium and large households nor to the type where most of the area leased in is accounted for by medium and large households and leased out by marginal and small households.

Table 2.

Distribution of Leased-in and Leased-out Land among Different Categories of Households in Agriculturally Developed and Backward States, 1992–2013 (%).

Category of households	Haryana
	Leased-in Land			Leased-out Land
	1992	2003	2013	1992	2003	2013
Landless	4.49	0.90	0.01	0.00	0.00	0.00
Sub-marginal	4.76	28.79	1.51	0.21	16.03	15.89
Marginal	6.34	11.16	13.82	5.75	6.30	0.70
Small	30.65	25.44	19.47	1.54	15.25	30.94
Medium	17.11	19.75	14.24	28.60	32.87	26.83
Large	36.65	13.96	50.95	63.89	29.55	25.64
All sizes	100.00	100.00	100.00	100.00	100.00	100.00
	Odisha
	Leased-in Land			Leased-out Land
	1992	2003	2013	1992	2003	2013
Landless	11.15	8.84	5.77	0.00	0.00	0.00
Sub-marginal	59.28	76.09	78.92	17.57	14.35	26.23
Marginal	18.72	12.69	6.19	11.82	24.93	34.53
Small	5.57	2.37	7.83	51.18	23.61	9.31
Medium	4.71	0.00	0.85	11.51	22.48	21.23
Large	0.56	0.00	0.44	7.92	14.63	8.70
All sizes	100.00	100.00	100.00	100.00	100.00	100.00

Source: NSSO unit-level data, 48th, 59th and 70th rounds of Land & Livestock Holdings.

Note: Landless who own less than 0.002 ha; sub-marginal between 0.002–0.5 ha; marginal 0.5–1 ha; small 1–2 ha; medium 2–4 ha and large above 4 ha.

H = Households, A = Area

The extent of reverse tenancy has also been examined in terms of concentration of land leased in and leased out among households at different levels of land ownership hierarchy in both the categories of states. Reverse tenancy in terms of concentration of leased-in and leased-out land implies concentration of leased-in land among households at top different levels and concentration of leased-out land among households at the bottom of the pyramid. We have, therefore, computed the concentration of leased-in and leased-out land at different levels of land ownership hierarchy. The results in this regard are presented in Table 3. As may be seen from the table, in agriculturally developed states, between 1992 and 2013, while the concentration of leased-in land declined continuously at the bottom 50%, it nearly doubled among the middle 30% households who accounted for nearly half of the total leased-in land. The concentration of leased-in land among households at top different levels, however, decreased during the period. On the other hand, changes in the concentration of leased-out land at different levels during the period show that while there was a small increase in the concentration of leased-out land among households at the top 20% who accounted for around 52% of the total leased-out land in 2013, the households at the bottom 50% and middle 50%–80% together accounted for around 48% of total leased-out land. Thus, in terms of changes in the concentration of leased in and leased-out land at different levels of ownership hierarchy, the results show that while most of the leased-in land is concentrated among households at the bottom 50% and middle 50%–30% levels, the leased-out land is concentrated among those who are at top different levels beginning with top 20% lending support to traditional pattern of tenancy.

Table 3.

Concentration of Leased-in and Leased-out Land at Different Levels of Land Ownership Hierarchy in Agriculturally Developed and Backward States, 1992–2013.

Year	Concentration of Leased-in Land at the						Concentration of Leased-out Land at the
	Top				Middle	Bottom	Top				Middle	Bottom
	1%	5%	10%	20%	50–80%	50%	1%	5%	10%	20%	50–80%	50%
	Haryana
1992	5.03	29.00	40.04	48.30	26.18	25.52	1.28	12.89	26.38	49.54	33.33	17.13
2003	7.55	23.43	32.86	55.44	22.25	21.51	14.44	28.57	42.73	66.32	27.95	5.73
2013	0.61	6.25	12.55	27.85	49.24	22.91	2.02	12.77	28.87	52.40	33.96	13.64
	Odisha
1992	1.56	9.35	17.59	31.90	53.93	14.17	4.90	11.04	19.33	43.77	36.49	20.04
2003	0.62	5.99	13.86	18.69	38.10	43.21	7.42	22.25	34.46	52.78	33.45	13.77
2013	2.66	9.54	14.19	22.87	14.57	62.56	8.88	27.06	35.56	49.35	32.15	18.50

Source: Authors’ own calculation based on data available in:

Report on Operational Land Holdings in India; 48th round 1991–92, NSS Report No. 407.

Report on Some Aspects of Operational Land Holdings in India; 59th round 2002–03, NSS Report No. 492.

Report on Household Ownership and Operational Holdings in India; 70th round 2012–13, NSS Report No. 571.

In agriculturally backward state, the concentration of the leased-in land among the households at the bottom 50% increased more than four times between 1992 and 2013, who in the later year accounted for more than three-fifths of the total leased-in land. Insofar as changes in the concentration of leased-out land are concerned, the table shows that around 50% of the total leased-out land was concentrated among households at top 20%.

Thus, in terms of concentration of leased-in and leased-out land, the tenancy relations in agriculturally backward state were of traditional type where most of the leased-in land is concentrated among households at lower levels of land ownership hierarchy and leased-out land among those at the top different levels.

In broad terms, these results suggest that, in agriculturally developed state of Haryana, the phenomenon of reverse tenancy in terms of the distribution of leased-in and leased-out land among households of different farm size categories is more pronounced. In comparison, in agriculturally backward states, the business of leasing in and leasing out land is broadly confined among households belonging to not very different size categories, that is, among sub-marginal, marginal and small households with important implications for negotiating the terms of tenancy contracts. In terms of concentration of leased-in and leased-out land at different levels of ownership hierarchy, the tenancy relations broadly conform to the traditional pattern in both the categories of states.

Determinants of Reverse Tenancy

As mentioned above, the reverse tenancy is a phenomenon where households of lower size categories lease out and those of higher size categories lease in land. As alluded to above, there could be multiple factors forcing households of lower size categories leasing out and those of higher size categories leasing in land. To quantify contribution of different factors affecting decision of different categories of households to lease in and lease out land, two logit regression models for each of the two states representing different levels of agricultural development were estimated. In the first model, dependent variable was medium and large farmers leasing in land assuming value 1 and all others taking value 0 while, in the second model, it was sub-marginal, marginal and small households leasing out land taking value 1 and all others taking vale 0. The choice of the independent variables was constrained by the availability of data in different NSS rounds. The variables considered for two models are household size, age of the head of the family, education of the head of the family, ownership of livestock, occupation, social category of a household and land use. Further, for both the states, the logit models are estimated separately for each of the three years in that the importance of different factors affecting decision of different categories of households to lease in and lease out land change over time because of changes in the rural economy in general and agricultural sector in particular. The results of the regression models are presented in Table 4 and Table 5.

Table 4.

Determinants of Reverse Tenancy in Haryana: Results of Logistic Regression Analysis, 1992–2013.

Explanatory Variables	Dependent Variable: Area Leased-in by Medium and Large Households = 1; Else = 0			Dependent Variable: Area Leased-out by Small, Sub-marginal and Marginal Households = 1; Else = 0
Explanatory Variables	1992	2003	2013	1992	2003	2013
Constant	–2.601***(–3.82)	–4.330***(–5.84)	–3.692***(–8.43)	1.130(1.21)	1.278(1.366)	6.626***(4.19)
Household size	0.101**(0.03)	0.127***(4.36)	0.110***(6.83)	–0.167**(–2.01)	–0.217**(–2.62)	–1.564***(–4.76)
Age (years)	0.013*(1.76)	0.021**(2.40)	0.007(1.36)	–0.007(–0.63)	0.023(1.15)	0.038**(2.08)
Education (years)	–0.049(–0.55)	0.110**(2.43)	0.074***(3.51)	0.123(0.83)	0.145(1.27)	5.677***(4.18)
Livestock: Yes = 1, else = 0	–	0.383(1.25)	0.213*(1.80)	–1.265**(–2.28)	–0.823(–1.26)	–
Self-employed in agriculture = 1, else = 0	–	1.055**(2.42)	–	1.445***(3.67)	–	–
Land used for food-grains = 1, else = 0	0.261(0.53)	0.228(1.08)	0.445*(1.76)	–	–	–
Pseudo R-squared	0.1306	0.2006	0.1840	0.3453	0.1605	0.3888
No. of observations	206	226	765	111	57	197

Source: Authors’ own calculation based on data available in:

Report on Operational Land Holdings in India; 48th round 1991–92, NSS Report No. 407.

Report on Some Aspects of Operational Land Holdings in India; 59th round 2002–03, NSS Report No. 492.

Report on Household Ownership and Operational Holdings in India; 70th round 2012–13, NSS Report No. 571.

Notes: *, ** and *** denote levels of significance at 10%, 5%and 1%respectively. Figures in parentheses are z values.

Table 5.

Determinants of Reverse Tenancy in Odisha: Results of Logistic Regression Analysis, 1992–2013.

Explanatory Variables	Dependent Variable: Area Leased-in by Medium and Large Households = 1; Else = 0			Dependent Variable: Area Leased-out by Small and Marginal Households = 1; Else = 0
Explanatory Variables	1992	2003	2013	1992	2003	2013
Constant	–5.699***(–5.59)	–18.935(–1.09)	–5.910***(–3.55)	2.559***(7.33)	7.523***(3.83)	2.317***(3.02)
Household size	0.138***(0.03)	0.073(0.18)	0.374***(5.80)	–0.132***(–5.02)	–0.491***(–3.20)	–0.067*(–1.67)
Age (years)	–0.014*(–1.17)	0.248(0.96)	–0.008(–0.35)	–0.003(–0.60)	0.013(–0.64)	–0.023**(–1.98)
Education (years)	0.341**(2.24)	0.546(0.52)	–0.091(–1.02)	–0.276***(–4.19)	–0.234**(–1.97)	0.085**(1.88)
Social category: SC/ST = 1; else = 0	0.596**(1.96)		0.364(0.72)	–	–	–0.038(–0.08)
Livestock: Yes = 1, else = 0	–	–	–	–1.052***(–6.56)	–	–
Self-employed in agriculture = 1, else = 0	2.252***(3.06)	–	0.681(0.86)	–0.485**(2.45)	–2.257***(–3.15)	–
Land use: Non-food-grains = 1, else = 0	–	–	0.891(0.81)	–	–	–
Pseudo R-squared	0.1133	0.2298	0.2995	0.2901	0.6282	0.1233
No. of observations	709	56	520	467	103	138

Source: Authors’ own calculation based on data available in:

Report on Operational Land Holdings in India; 48th round 1991–92, NSS Report No. 407.

Report on Some Aspects of Operational Land Holdings in India; 59th round 2002–03, NSS Report No. 492.

Report on Household Ownership and Operational Holdings in India; 70th round 2012–13, NSS Report No. 571.

Notes: *, ** and *** denote levels of significance at 10%, 5%and 1%respectively. Figures in parentheses are z values.

The results of the logit regressions for Haryana, an agriculturally developed state, are presented in Table 4. A perusal of the table shows that the variables such as household size, age of the head of the family, education of the head of the family, ownership of livestock, self-employment in agriculture and land used for food grains affected the probability of medium and large households leasing in land. In all the three years, most of these variables had expected signs. The notable exceptions to this broad pattern were variables such as education of the head of the family and land used for foodgrains in 1992, ownership of livestock and land used for foodgrains in 2003 and age of the head of the family in 2013. These variables in these years either did not have expected signs or the regression coefficients were statistically insignificant. Regarding factors affecting the probability of sub-marginal, marginal and small households leasing out land, the table shows that in 1992 household size and ownership of livestock, as expected, had negative signs and were statistically significant suggesting that a household in these size categories who has large household size and own livestock will not be inclined to lease out his land. However, contrary to expectations, the results show that a household self-employed in agriculture will have a higher probability of leasing out land which is difficult to explain. In 2003, only one variable, that is, the size of the household was statistically significant and had expected sign implying that a household with large size will have a negative probability of leasing out his land. All other variables, though had expected signs, were statistically insignificant. And, in 2013, in addition to household size, age of the head of the family and education of the head of family were statistically significant with expected signs implying that that probability of sub-marginal, marginal and small households leasing out land increases with increase in age and education of the head of the family.

The results of the logit regression models for agriculturally backward state of Odisha are presented in Table 5. First, we consider the results of the model quantifying the effect of variables on the probability of a mediumnd large households leasing in land. As may be seen from the table, in 1992, the variables such as household size, age of the head of the family, education of the head of the family, social category of the household and self-employment in agriculture were statistically significant and had positive effect on the probability of such households leasing out land except that age of the head of the family had negative effect. However, in 2003, though all these variables continued to have had a positive effect on the probability of these households leasing in land, the regression coefficients associated with these variables were statistically insignificant. Similarly, in 2013, household size only had a positive and statistically significant effect on the probability of such households leasing in land. All other variables were statistically insignificant, and some of these also did not have expected signs. The results of the logit regression model quantifying the effect of variables on the probability of sub-marginal, marginal and small households leasing out land show that, in all the 3 years, four variables, namely, household size, education of the head of the family, ownership of livestock and self-employment in agriculture, had negative and statistically significant effect on the probability of such households leasing out land. It is, however, important to mention that among these variables, the signs of the coefficients associated with education of the head of the family in 1992 and 2003 had negative signs, implying that more educated these households are higher the probability of their not leasing out land which is contrary to expectations.

Summary and Conclusion

Indian agriculture since the 1980s and the 1990s has witnessed huge transformation in terms of adoption of new agricultural technology across crops and regions, demand-driven diversification in cropping pattern and commercialisation, increase in the migration of rural households to urban areas in response to increase in rural non-farm employment opportunities and increase in the disparities between wage earnings in the agricultural and non-agricultural sectors resulting in increase in the ratio of income per non-agricultural worker to farm income per cultivator. In more recent times, Indian agriculture is also witnessing growing agrarian distress manifested in rising cost of production, falling output prices, falling incomes, increasing indebtedness among farmers and, ultimately, increase in the farmers suicides. These changes in agricultural and non-agricultural sectors are encouraging households of lower farm size categories such as sub-marginal, marginal and small to lease out their land and medium and large households lease in land, giving rise to a phenomenon popularly known as reverse tenancy. More recent studies have reported that these tendencies are more pronounced in agriculturally developed regions as compared to agriculturally backward regions. It is against this background that the present study examines the extent of reverse tenancy and the factors determining the extent of reverse tenancy in agriculturally developed and backward states using NSS unit-level data from different rounds.

The findings of the study throw up very interesting and mixed patterns. For example, it shows that, in agriculturally developed state, in terms of the distribution of leased-in and leased-out land among different size categories of households, nearly two-thirds of the total leased-in land is accounted for by medium and large households who also contribute more than half of the total leased-out land. It is, however, equally important to mention that around 47% of the total leased-out land is accounted for by sub-marginal, marginal and small households. These results together point towards the prevalence of reverse tenancy in agriculturally developed state. However, in terms of changes in the concentration of leased-in and leased-out land at different levels of land ownership hierarchy, the results show that, even in agriculturally developed state, the tenancy relations are of traditional type in that while most of the leased-in land is concentrated among households at the bottom 50% and middle 50%–30% levels, the leased-out land is concentrated among those who are at top 20%, 10%, 5% and 1% levels. Insofar as tenancy relations in agriculturally backward states are concerned, these conform neither to the traditional type where most of the leased-in land is accounted for by households of lower size categories and leased out by those in higher size categories nor to the type where most of the leased-in land is accounted for by medium and large households and leased out by sub-marginal, marginal and small households. These results suggest that the leasing in and leasing out of land in agricultural backward states broadly takes place among households belonging to not very different size categories of landholdings. However, going by the concentration of leased-in and leased-out land at different levels of ownership hierarchy, the results show the prevalence of traditional tenancy relations where most of leased-in land is concentrated among households at the bottom 50% and middle 30% and leased-out land among households at top different levels. The results of the logit regression further show that factors such as household size, age of the head of family, education of the head of the family, self-employment in agriculture, ownership of livestock and land use affect the probability of medium and large households leasing in and sub-marginal, marginal and small households leasing out land though the nature of their effect and statistical significance vary among agriculturally developed and backward states and also over the years.

On the basis of the findings of the study, the authors suggest that the centre should persuade the state governments to amend their tenancy laws to allow leasing in and leasing out land on the basis of the provisions suggested the Model Agricultural Land Leasing Act, 2016 prepared and approved by the NITI Aayog. The land markets in India have remained sluggish as the farmers are not willing to sell their land because of their attachment to land and lack of social security. They are also reluctant to lease out their land fearing loss of land ownership and instead prefer to keep it fellow in the event of them not being able to cultivate it, causing a wastage and degradation of scarce land resource. Therefore, legalising lease market would go a long way not only to activate the land market and allocate scarce land resources to more productive uses but also to impart much needed flexibility and occupational mobility in the rural economy. This will, inter alia, lead to improvement in agricultural efficiency, equity, access to land by landless, marginal and small households and encourage all categories of farmers to participate in the lease market depending upon their resource endowment, availability of alternative employment opportunities, their risk bearing capabilities and so on. Legalising leasing in and leasing out land will also enable the lessees to claim the status of the farmers and benefit from different schemes which are available to the farmers and the lessors to resume their land without any encumbrances after the expiry of lease.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Note

ORCID iD

Shakir Hussain Malik

References

Bellemare

M. F.

(2007). Testing between competing theories of reverse share tenancy. Terry Sanford Institute of Public Policy, Duke University .

Bharadwaj

, & Das

P. K.

(1975). Tenurial conditions and mode exploitation: A study of some villages in Orissa. Economic and Political Weekly , 10(5–7), 221–240.

Carnham

, Luther

H. A.

, & Wilkes

J. O.

(1969). Applied numerical methods . Krieger Publishing Company.

Chand

, Srivastava

S. K.

, & Singh

(2017). Changing structure of rural economy of India: Implications for employment and growth . National Institution for Transforming India.

Dasgupta

(1984). Sharecropping in West Bengal: From independence to operation Barga. Economic and Political Weekly, 19(26), A85–A87, A89–A96.

Iyer

G. K.

(2009). Globalization and contract farming: The experience of some states . https://books.google.co.in/books?hl=en&lr=&id=jJgfNVjZ8koC&oi=fnd&pg=PA399&ots=6RAyPeTSVB&sig=-3VNwAg2gqIGfN8ZsRr4JHdiTAI#v=onepage&q&f=false

Jodha

N. S.

(1981). Agricultural tenancy: Fresh evidence from dryland areas in India. Economic and Political Weekly , 16(52), A118–128.

Malik

S. H.

(2019). Tenancy relations in transition: An empirical study in agriculturally developed and underdeveloped states [PhD Thesis]. Central University of Himachal Pradesh, India.

Murty

C. S.

(2004). Large farmers in lease market: Are marginal farmers affected? Economic and Political Weekly , 39(29), 3270–3279.

10.

Nadkarni

M. V.

(1976). Tenants from the dominant class: A developing contradiction in land reforms. Economic and Political Weekly , 11(52), A137–146.

11.

Papola

T. S.

(2014). Economic diversification and labour market dynamics in rural India. Indian Journal of Agricultural Marketing , 28(3), 18–33.

12.

Rao

V. M.

(1974). Village lease market for agricultural land: Some approaches for analysis. Economic and Political Weekly , 9(26), A55–A62.

13.

Ray

C. N.

(2016). Agricultural tenancy in some irrigation command areas of North Gujarat. Journal of Land and Rural Studies , 4(2), 172–187.

14.

Ray

(1978). The small lessor and the big lessee: Evidence from West Bengal. Economic and Political Weekly , 13(51–52), A119–124.

15.

Sharma

H. R.

(2000). Tenancy relations in rural India: A temporal and cross-sectional analysis. Indian Journal of Agricultural Economics , 55(3), 295–307.

16.

Sharma

H. R.

(2005). Rapporteur’s report on changing agrarian relationships in rural areas. Indian Journal of Agricultural Economics , 60(3), 561–570.

17.

Sharma

H. R.

(2006). Liberalising the lease market. Economic and Political Weekly , 41(8), 696–698.

18.

Sharma

H. R.

, & Malik

S. H.

(2019). Changing agrarian structure in rural India, 1953-54 to 2012-13: Evidence from NSS Data. Indian Journal of Agricultural Economics , 74(4), 506–526.

19.

Shergill

H. S.

(1989). Agrarian structure as a factor in rural poverty: Some cross-section evidence. Economic and Political Weekly , 24(12), 9–12.

20.

Singh

C. P.

, & Sharma

H. I.

(2013). Peasants of Manipur: Agrarian change, land tenure and emerging patterns of re-peasantisation in India. Agrarian South: Journal of Political Economy , 2(3), 291–314.

21.

Singh

(1989). Reverse tenancy in Punjab agriculture: Impact of technological change. Economic and Political Weekly , 24(25), 86–92.

22.

Swain

(1999). Agricultural tenancy and interlinked transactions: I: Neoclassical and Marxist approaches. Economic and Political Weekly , 34(37), 2657–2666.

23.

Vijay

(2012). Structural retrogression in Indian agriculture and the rise of ‘New Landlords’: An empirical exercise. Economic and Political Weekly , 47(5), 37–45.

24.

Vijay

, & Sreenivasulu

(2013). Agrarian structure and land lease arrangements: An investigation in nine villages in Andhra Pradesh. Economic and Political Weekly , 48(26–27), 42–49.

25.

Vyas

V. S.

(1970). Tenancy in a dynamic setting. Economic and Political Weekly , 5(26), A73–A80.