Land per capita,year-to-year variability and uncertainty of return on investment affect household income from crop production: Evidence from India

Abstract

Agriculture's potential to reduce poverty assumes that income from crops can significantly increase per capita income of small farm households. Using nationally representative data on the value and variable costs of production of 10 crops (rice, maize, blackgram, chickpea, pigeonpea, mustard, groundnut, soybean, sunflower, sesamum) commonly grown in India, we calculated the mean benefit: cost (B:C) ratios and profit per hectare for each crop in seven three-year periods between 2000 and 2020. All crops were profitable (B:C > 1) in all periods but B:C varied widely and unpredictably, both between crops and with time. We then calculated, for the maximum and minimum observed national level values of B:C for each crop, the relation between household land per capita (LPC) over the range 0–1.0 hectares per person, and the crop personal daily income (CPDI) in PPP $ per person per day that would be generated by growing that crop. Only three out of a possible 100 combinations of crop and LPC (chickpea, 2005–2008, LPC = 1.0 and mustard (2008–2011, LPC = 0.9 or 1.0) generated CPDI values greater than a poverty line of $2.15 per person per day. A large survey confirmed that 96.9% of farming households in India had LPC values of 1.0 hectares per person or less. The majority of farming households could not have generated enough income from growing any of these crops during this twenty-year period to reach a poverty line of $2.15 per person per day. Most farms in India do not have enough land to benefit much from intensifying their crop production. Such small incentives and uncertainty about expected returns on investment (and increased exposure to risk) might contribute to low adoption of best practices, and more in-depth studies of rural household livelihood portfolios are needed to clarify that assumption.

Keywords

Benefit cost ratios historic variability personal daily income poverty line small farms

Introduction

Estimates suggest there are very many smallholder farmers in the world (Giller et al., 2021; Lowder et al., 2016, 2021; Lowder et al., 2025). These rural households are the target of many development efforts – to increase yields (Global Yield Gap Atlas, 2025; Van Ittersum et al., 2016) and so improve local, national and global food security (Godfray et al., 2010; Godfray and Garnett, 2014), reduce the impact of farming on the environment (United Nations, 2015), and to use crop and livestock production to move out of poverty (CGIAR, 2005; Harris and Orr, 2014).

This study focuses on crop production. To grow crops, to feed their families directly or to be sold to provide income, smallholder farming households must invest some resources. Those resources include time and effort, money (derived from previous farming, other income, savings, or credit) to buy inputs such as seeds, fertilisers, pesticides, tools, and perhaps additional labour. The other necessary resource is land, which may be owned, leased, shared or allocated by the community. By definition, smallholder households do not operate very much land.

In comparison with promoting higher yields, increasing the economic returns from smallholder agriculture has received less emphasis by the research and development community. In general, increasing yields involves increasing investment, and so putting more resources at risk if things go wrong. With limited resources, smallholder rural households need to feel confident that their investment in cropping will pay off reliably and that the benefit will be worth the investment. One easily appreciated measure of return on investment in crop production is the benefit: cost ratio (B:C), defined here in its simplest form as the gross value of all crop outputs divided by the variable costs of production. Values of B:C greater than 1 represent a profit whereas values less than 1 mean that the farmer has made a loss.

At first glance, investing in growing crops is a great way to make money. The literature is full of reports of technologies with B:C ratios greater than one, that is, they are profitable. In fact, larger values for B:C are commonly claimed (see a review by Rosegrant et al., 2023). For instance, almost all the improved rainfed cropping technologies surveyed by Harris and Orr (2014) had B:C ratios (calculated using costs including the imputed value of family labour) greater than two and some had values greater than three or even more. More recently, many authors, for example, Cartsburg et al. (2024), Sanchez et al. (2022), and Gathala et al. (2020) have reported similarly large returns on investment. So, if a rural household had access to some land and was prepared to spend (in money and in kind), say, a thousand dollars they could get an additional thousand dollars in a season (or year, depending on the environment and how many seasons per year it allowed) by engaging in a cropping enterprise with a B:C ratio of 2. Any bank or savings institution that offered such an unprecedented interest rate of 100% on investment would be inundated with investors. Unfortunately, smallholder rural households do not have much land, nor do they generally have much money to invest. Any money they do have is not all ‘spare’ and available for investment; some must be spent on food, clothing and other necessities. To increase adoption, interventions and innovations should therefore not only increase yields but also reduce input requirements per unit of output of those factors which are scarce or expensive (Binswanger, 1986).

India has very many farming households with small amounts of land and the country also has available large amounts of data describing those households and how they grow crops. In this paper we use those data to demonstrate that:

With some exceptions, most cropping enterprises are profitable, that is, have B:C ratios > 1.

B:C ratios of a range of crops vary considerably and unpredictably between seasons and between years.

Even cropping enterprises with high B:C values only generate small benefits per person for members of households with small amounts of land per capita.

To generate significant incomes, the investment required by smallholder households is unfeasibly high.

Our work complements prior studies on living income potential (or lack thereof) of agricultural investments in India. These include Chand et al. (2011), Chand et al. (2015), Gaurav and Mishra (2015), Reddy et al. (2020), and Singh et al. (2025). Despite not focusing on livestock, our study also complements studies on the puzzle of low economic returns in the livestock sector, especially dairy cattle in India (e.g. Attanasio and Augsburg, 2017; Gehrke and Grimm, 2018).

Materials and methods

Data

We use data from two sources: (i) the Government of India's Commission of Agricultural Costs and Prices (CACP) data summarised by Johnson and Singh (2022) and (ii) the National Sample Survey (NSS) 77^th Round Situation Assessment of Agricultural Households and Land and Holdings of Households in Rural India. Johnson and Singh (2022) reported trends in costs of cultivation, prices, and incomes for seven three-year periods during 2000–2001 to 2019–2020 for 10 agricultural crops grown in India: rice; maize; blackgram; chickpea; pigeonpea; mustard; groundnut; soybean; sunflower; and sesamum. We report much of the results for these 10 crops, but we also analysed the original CACP data for 17 other crops including wheat and sugarcane for the period 2011–2020 with similar general results (Table S2).

From the Tables in their report, we extracted costs per hectare (A2 + FL) and the gross value of output per hectare (GVO) for all crops and periods. As defined by GOI (2008), A2 + FL represents all paid-out costs such as for seeds, fertilisers, manure, insecticides, etc., hired human-, machine-, and animal labour, interest on working capital, rent paid for leased-in land, and depreciation and maintenance expenditure incurred for owned machinery, and the imputed value of family labour. GVO is calculated as the sum of the values of the main product and any by-products. Rupees were converted to Purchasing Power Parity US dollars using conversion factors representative of each three-year period.

The Situation Assessment of Agricultural Households and Land and Holdings of Households in Rural India, India, NSS 77^th Round (National Sample Survey, 2019) provides a nationally representative sample of Indian farming households. We used two variables, household size and farm size to calculate land per capita (LPC) and its distribution.

Stylised identities

For each of the 10 crops, we used the maximum and minimum values of B:C during the 20-year period (‘bounds analysis’, Pardey et al., 2014) to calculate the personal daily income that could be derived from crop production (CPDI, in PPP $ per person per day) of members of farming households with land per capita ratios ranging from 0.1 to 1.0 hectares per person. We used the relation

C P D I = (N e t p r o f i t p e r h e c t a r e \times L P C) / 365

(1)

This stylised identity has previously been used by Harris and Orr (2014). It is also similar to the physical productivity identity (e.g. Gollin, 2018; Hayami and Ruttan, 1985) that states that labour productivity (output per labour) is a product of land productivity (output per land) and land per capita.

Starting from this measure, we can then compute the two measures that a farmer can change in the short run to improve CPDI: (i) investments other than land that can raise the productivity and therefore profits per unit of land, and (ii) the amount of land per capita required to achieve a particular threshold.

Using the same maximum and minimum cases for crops and the same range of LPC, we calculated the amount of investment per household that would be required to generate a CPDI value of PPP $2.15/person/day, that is, to reach the international poverty line as updated by the World Bank Group in 2022, using the relation

I n v e s t m e n t r e q u i r e d p e r h o u s e h o l d p e r y e a r = o r i g i n a l c o s t s \times ((2.15 \times 365) / (o r i g i n a l p r o f i t \times L P C))

(2)

and the land required if that investment were to be applied to the mean farm using

\begin{aligned} L a n d r e q u i r e d \\ = (2.15 \times m e a n H H s i z e \times 365) \\ / p r o f i t p e r h o u s e h o l d w i t h m e a n f a r m s i z e \end{aligned}

(3)

assuming each crop is only grown once per year.

In applying equations (1) to (3), we recognise that we describe a somewhat artificial situation, in that farmers often do not grow any of the 10 crops on all of their land in any season, nor are they restricted to these ten crops. Nor are all farmers restricted to one growing season per year if rainfall or irrigation is available and temperatures are suitable for supporting plant growth, although it should be borne in mind that such additional crops require additional costs of a similar magnitude to those encountered in this study and would be exposed to similar levels of risk and uncertainty of outcome. We contend that our approach is appropriate to demonstrate the year-to-year volatility and unpredictability of the profit, and consequent income per person, to be gained from crop production in India during this twenty-year period.

Results and discussion

Benefit: cost ratios of crop production

Figure 1 shows mean values of B:C ratio for 10 common crops during seven three-year periods between 2000 and 2020. Means, medians and estimates of intertemporal variability for each crop are summarised in Table S1. We report much of the results for these 10 crops, but we also analysed the original CACP data for 17 other crops including wheat and sugarcane for the period 2011–2020 with similar general results (Table S2). These estimates are consistent with prior evidence from Chand et al. (2015) who documented B:C values that ranged from 1.31 to 1.7 between 1983–1984 and 2011–2012. All crop x period combinations are profitable, that is, B:C is >1, but some crops are consistently more profitable than others in all periods. However, there is a large amount of variation in the B:C value of all crops from period to period. We assume that B:C is the resultant of various interactions between land quality, weather, management, policy, markets, prices, etc., and so is impossible to predict in advance of choosing a crop and planting it. In that sense, crop profitability is uncertain and hence investment in crop production carries an unknown degree of risk. This is consistent with evidence by Agarwal and Agrawal (2017) that almost 40% of farmers in India expressed that they don’t like farming because of low profitability (60%) seconded by risks (19%).

Figure 1.

Benefit: cost ratios (gross income per hectare minus variable costs per hectare including imputed cost of family labour) of 10 crops in India between 2000 and 2020. Data from: Trends in costs and incomes from crop production in India: analysis of costs of cultivation data, 2000–2001 to 2019–2020. A project report prepared by the Foundation for Agrarian Studies for International Rice Research Institute, November, 2022.

Table 1 shows the maximum and minimum B:C ratios of the ten crops in Figure 1, together with the associated costs and benefits, both in Indian Rupees and in Purchasing Power Parity dollars calculated using the appropriate conversion factor for the period. Supplemental Table S1 shows the corresponding mean and median values. Both costs and gross incomes per hectare vary markedly between the maximum and minimum periods and the differences in levels of profitability are substantial for all crops, ranging from a 17% reduction (defined as the difference as a proportion of the maximum) in rice to a 51% reduction in blackgram. It should be noted that the returns on investment and profits in Table 1 (and hence also the values of CPDI in Table 2) do not consider any fixed costs or taxes that might have been applicable to farmers’ circumstances, so are almost certainly overestimates of actual profitability.

Table 1.

Maximum and minimum values (see Figure 1) of benefit: cost ratios (gross income per hectare divided by variable costs per hectare including imputed cost of family labour) of 10 crops in India between 2000 and 2020.

Crop	Year	PPP conversion factor	Variable cost, Rs/ha	Gross income, Rs/ha	Profit, Rs/ha	Variable cost, PPP^a $/ha	Gross income, PPP $/ha	Profit, PPP $/ha	B:C ratio
Rice max	2008–2011	13.14	13,420	22,838	9418	1021	1738	717	1.70
Rice min	2014–2017	18.42	15,627	21,976	6349	848	1193	345	1.41
Maize max	2008–2011	13.14	9777	15,723	5946	744	1197	453	1.61
Maize min	2000–2002	10.06	8201	8931	730	815	888	73	1.09
Blackgram max	2005–2008	10.95	5213	11,489	6276	476	1049	573	2.20
Blackgram min	2017–2020	19.75	7621	8228	607	386	417	31	1.08
Chickpea max	2005–2008	10.95	6735	16,100	9365	615	1470	855	2.39
Chickpea min	2017–2020	19.75	9758	16,102	6344	494	815	321	1.65
Pigeonpea max	2008–2011	13.14	9679	19,610	9931	737	1492	756	2.03
Pigeonpea min	2017–2020	19.75	12,589	18,020	5431	637	912	275	1.43
Mustard max	2008–2011	13.14	7282	19,267	11,985	554	1466	912	2.65
Mustard min	2000–2002	10.06	8028	14,998	6970	798	1491	693	1.87
Groundnut max	2011–2014	15.44	17,988	28,062	10,074	1165	1817	652	1.56
Groundnut min	2000–2002	10.06	10,985	14,076	3091	1092	1399	307	1.28
Soybean max	2011–2014	15.44	9085	15,883	6798	588	1029	440	1.75
Soybean min	2014–2017	18.42	10,365	12,366	2001	563	671	109	1.19
Sunflower max	2005–2008	10.95	5732	8821	3089	523	806	282	1.54
Sunflower min	2000–2002	10.06	6422	6752	330	638	671	33	1.05
Sesamum max	2008–2011	13.14	5410	11,135	5725	412	847	436	2.06
Sesamum min	2017–2020	19.75	7904	10,481	2577	400	531	130	1.33

Data from: Trends in costs and incomes from crop production in India: analysis of costs of cultivation data, 2000–2001 to 2019–2020. A project report prepared by the Foundation for Agrarian Studies for International Rice Research Institute, November, 2022.

Purchasing Power Parity conversion factors from https://www.indexmundi.com/facts/india/ppp-conversion-factor#PA.NUS.PRVT.PP.

Table 2.

Personal daily income (PPP $/person/day) from crops at the maximum and minimum values of B:C ratio (2000 to 2020) by land per capita (ha/person). Only shaded values exceed $2.15/person/day.

				Land per capita (ha/person)
Crop	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9	1
Rice max	0.20	0.39	0.59	0.79	0.98	1.18	1.37	1.57	1.77	1.96
Rice min	0.09	0.19	0.28	0.38	0.47	0.57	0.66	0.76	0.85	0.94
Maize max	0.12	0.25	0.37	0.50	0.62	0.74	0.87	0.99	1.12	1.24
Maize min	0.02	0.04	0.06	0.08	0.10	0.12	0.14	0.16	0.18	0.20
Blackgram max	0.16	0.31	0.47	0.63	0.79	0.94	1.10	1.26	1.41	1.57
Blackgram min	0.01	0.02	0.03	0.03	0.04	0.05	0.06	0.07	0.08	0.08
Chickpea max	0.23	0.47	0.70	0.94	1.17	1.41	1.64	1.87	2.11	2.34
Chickpea min	0.09	0.18	0.26	0.35	0.44	0.53	0.62	0.70	0.79	0.88
Pigeonpea max	0.21	0.41	0.62	0.83	1.04	1.24	1.45	1.66	1.86	2.07
Pigeonpea min	0.08	0.15	0.23	0.30	0.38	0.45	0.53	0.60	0.68	0.75
Mustard max	0.25	0.50	0.75	1.00	1.25	1.50	1.75	2.00	2.25	2.50
Mustard min	0.19	0.38	0.57	0.76	0.95	1.14	1.33	1.52	1.71	1.90
Groundnut max	0.18	0.36	0.54	0.72	0.89	1.07	1.25	1.43	1.61	1.79
Groundnut min	0.08	0.17	0.25	0.34	0.42	0.51	0.59	0.67	0.76	0.84
Soybean max	0.12	0.24	0.36	0.48	0.60	0.72	0.84	0.97	1.09	1.21
Soybean min	0.03	0.06	0.09	0.12	0.15	0.18	0.21	0.24	0.27	0.30
Sunflower max	0.08	0.15	0.23	0.31	0.39	0.46	0.54	0.62	0.70	0.77
Sunflower min	0.01	0.02	0.03	0.04	0.04	0.05	0.06	0.07	0.08	0.09
Sesamum max	0.12	0.24	0.36	0.48	0.60	0.72	0.84	0.95	1.07	1.19
Sesamum min	0.04	0.07	0.11	0.14	0.18	0.21	0.25	0.29	0.32	0.36

It would be appropriate to ask if these low levels of profitability and high levels of variability across some of the crops could be resolved by reallocating land and other resources (e.g. irrigation, fertilisers) towards other crops. Variants of this proposal have been advocated based on long-term experimental datasets (e.g. Nayak et al., 2025), and in crop land use optimisation models (e.g. Pal et al., 2023; Wei et al., 2025). These normative assessments however have been superseded by real world assessments from randomised control trials (RCTs) in which for example Lybbert et al. (2024) found that farmers did not consistently switch to pulses even after receiving support. In terms of risk mitigation, there is also evidence from surveys that there is usually positive correlation of farm income sources which reduces the potential of crop diversification as a risk mitigation measure (Gaurav and Mishra 2015).

The values of CPDI calculated (equation (1)) using the profit data from Table 1 are shown in Table 2 for a range of LPC values between 0.1 and 1.0. CPDI values are very small, with only chickpea during the period 2005–2008 (for households with LPC values of 1) and mustard during the period 2008–2011 (for households with 0.9 or more hectares per capita) generating enough income to reach the internationally recognised poverty threshold of $2.15 per person per day. No other crop x household combinations could do this over the range of LPC considered.

In order to determine if findings in Table 2 are applicable to agricultural households in India, we used the data from National Sample Survey (2019) to calculate the distribution of land per capita of 30,719 farming households representative of India in general (Figure 2). The distribution is heavily skewed towards smaller values of LPC, with a median of 0.135 hectares per person and a mean of 0.226 hectares per person. Only 3.1% of households had LPC values greater than 0.9, so the calculations in Table 2 span the appropriate range of LPC in India. Thus, growing any of these ten crops was unlikely to have contributed enough value to cross the $2.15 poverty line for the vast majority of the farming households in the survey and, by implication, in India. Chand et al. (2011) also found that ‘tiny holdings below 0.8ha do not generate enough income to keep a farm family out of poverty despite high productivity. Nearly three-fourths of small farmers in India fall under poverty if they do not get income from non-farm sources’.

Figure 2.

Distribution (%) of land per capita of farming households in India. Values based on utilised land in kharif season, 2019. Data from National Sample Survey (2019). N = 30,719 farming households in 28 States and 8 Union Territories.

However, it is important here to recognise the importance of considering LPC rather than farm size per se. For instance, a household with an LPC value of 0.1 could comprise 10 people farming one ha, 20 people farming two hectares or 30 people farming three hectares, etc. For a crop with a B:C value of 1.7, the investment required per hectare to generate $2.15/p/d would be $11,211 for all three farms but would require $22,421 and $33,632 per household.

Required household investments to generate even a basic level of income

For any given value of B:C ratio, increasing the level of investment should result in a proportional increase in profit. The amounts required for households to generate $2.15 per person per day are shown (calculated using equation (2)) for the ten crops over the range of 0–1 ha per person in Table 3. For most values of LPC the amounts required are very large, particularly at minimum levels of crop performance with small, yet still profitable, B:C ratios and particularly for large households. This is illustrated in Figure 3 for household sizes ranging from 2 to 8. The solid line represents the mean household size, 5.13, calculated from the survey (National Sample Survey, 2019).

Figure 3.

Relation between B:C ratio, household size and the annual investment required to generate $2.15 per person per day from crop production. Mean sample household size = 5.13 (solid line).

Table 3.

The investment required (PPP $/household) to generate a personal daily income of PPP $2.15/person/day from any of 10 crops given the maximum and minimum values of B:C ratio for each crop (2000 to 2020) by land per capita (ha/person).

				Land per capita (ha/person)
Crop	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9	1
Rice max	11,182	5591	3727	2796	2236	1864	1597	1398	1242	1118
Rice min	19,315	9658	6438	4829	3863	3219	2759	2414	2146	1932
Maize max	12,904	6452	4301	3226	2581	2151	1843	1613	1434	1290
Maize min	88,161	44,080	29,387	22,040	17,632	14,693	12,594	11,020	9796	8816
Blackgram max	6518	3259	2173	1630	1304	1086	931	815	724	652
Blackgram min	98,527	49,263	32,842	24,632	19,705	16,421	14,075	12,316	10,947	9853
Chickpea max	5644	2822	1881	1411	1129	941	806	705	627	564
Chickpea min	12,071	6035	4024	3018	2414	2012	1724	1509	1341	1207
Pigeonpea max	7648	3824	2549	1912	1530	1275	1093	956	850	765
Pigeonpea min	18,190	9095	6063	4548	3638	3032	2599	2274	2021	1819
Mustard max	4768	2384	1589	1192	954	795	681	596	530	477
Mustard min	9039	4519	3013	2260	1808	1506	1291	1130	1004	904
Groundnut max	14,012	7006	4671	3503	2802	2335	2002	1752	1557	1401
Groundnut min	27,889	13,944	9296	6972	5578	4648	3984	3486	3099	2789
Soybean max	10,488	5244	3496	2622	2098	1748	1498	1311	1165	1049
Soybean min	40,649	20,325	13,550	10,162	8130	6775	5807	5081	4517	4065
Sunflower max	14,562	7281	4854	3640	2912	2427	2080	1820	1618	1456
Sunflower min	152,717	76,359	50,906	38,179	30,543	25,453	21,817	19,090	16,969	15,272
Sesamum max	7416	3708	2472	1854	1483	1236	1059	927	824	742
Sesamum min	24,069	12,035	8023	6017	4814	4012	3438	3009	2674	2407

The costs associated with generating $2.15 per person per day are far larger than the variable costs per hectare for the 10 crops that range from $386 per hectare (blackgram in 2017–2020) to $1165 per hectare (groundnut in 2011–2014), a range that is indicative of the amounts that households may be willing and able to pay. In contrast, a household with the sample mean size of 5.13 would have to invest from $1342 (for an enterprise with B:C = 4) to $16,104 (if B:C were only 1.25). Supplemental Table S3 shows the multiples of current investment required. Investing such large amounts in enterprises that, as we have seen in Figure 1 and Table 1 are inherently unpredictable, is not rational, feasible, or possible for resource-poor households, particularly since production curves are fundamentally asymptotic (i.e. diminishing returns to inputs) and investment beyond a certain point, that is, lavishing very large amounts of resources on small plots, is a waste and would result in large losses. Even modern best-practice cropping technologies with large values of B:C seldom require annual investment of more than $500 per hectare (e.g. Nayak et al., 2022 for maize, one season); $1000 per hectare (e.g. Jat et al., 2014 and Kakraliya et al., 2018 for the two-season rice/wheat system); $1600 (e.g. Kader et al., 2022 and Laik et al., 2014 for the three-season rice/wheat/mungbean system).

Smallholder farmers have two options for increasing their income from crop production – increase the B:C ratio, or increase their value of LPC, that is, operate more land, although these options are not mutually exclusive and can be implemented together in various combinations. Increasing the B:C ratio is possible and there are many improved technologies available to do so. Harris and Orr (2014) listed a range of improved rainfed enterprises with much larger B:C values than those of ‘farmer practice’, with several values above 4. More recent literature abounds with examples of modern crop production technologies with B:C values greater than farmer practice and values of 3 or 4 are commonly reported by researchers (e.g. Singh, 2021 for blackgram; Sharma and Singh, 2024 for chickpea; Tripathy et al., 2023 for mustard and chickpea; Singh et al., 2020 for sesamum; Sandhya and Singh, 2023 for sunflower). Since B:C values in Table 1 range from only 1.08 to 2.65 there would seem to be considerable additional benefit for farm households if they adopted best practices. However, it should be noted that we do not know the proportion of the values calculated in this paper from Johnson and Singh (2022) that represent ‘farmer practice’ rather than reflecting some degree of adoption of ‘improved practice’. The amount of ‘improvement’ over the situation summarised in Tables 1 and 2 that is possible, therefore, is uncertain. There is however a dearth of studies that estimate not only yield gaps but also profit gaps that can help to demonstrate the possibilities (e.g. Manjunatha et al., 2013).

If adopting best practices and pouring huge amounts of resources into cultivating small pieces of land makes little sense, could smallholder households increase their incomes meaningfully from crop production by operating more land? Figure 4 shows, for the mean (0.98 ha, 5.13 people) household, the farm size (calculated using equation (3)) that would be required to generate $2.15 per person per day by growing each of the 10 crops with minimum and maximum profitability. All cases would require substantial investment (not estimated here) in renting in or buying additional land. In addition to that investment to gain access to extra land, growing those crops on the larger farms would require scaled-up variable costs to achieve the same pro-rata outputs. Table 4 shows how much that extra investment would be for the mean household.

Figure 4.

Farm size (ha) required by the sample mean household (5.15 people, 0.98 ha) to generate a personal daily income of PPP $2.15/person/day from 10 crops given the maximum and minimum values of B:C ratio for each crop over the period 2000 to 2020 in India.

Table 4.

Original variable costs (PPP $/farm/year, assuming one crop per year) and new costs (over and above cost of buying or renting in new land) required to generate PPP $2.15/person/day for the average farm household in India growing any of 10 crops, calculated using the maximum and minimum benefit: cost ratios for each crop between 2000 and 2020.

	Mean farm
Crop	Original	New
Rice max	997	5740
Rice min	828	9914
Maize max	726	6623
Maize min	796	45,252
Blackgram max	465	3346
Blackgram min	377	50,573
Chickpea max	601	2897
Chickpea min	482	6196
Pigeonpea max	719	3926
Pigeonpea min	622	9337
Mustard max	541	2447
Mustard min	779	4639
Groundnut max	1137	7192
Groundnut min	1066	14,315
Soybean max	574	5383
Soybean min	549	20,865
Sunflower max	511	7474
Sunflower min	623	78,388
Sesamum max	402	3806
Sesamum min	391	12,355

Limitations and future research

There are limitations to the analytical approach we have taken in this paper. Mean and median values can obscure the best performances (but also the worst ones) and there may have been a significant proportion of farm households that grew these crops more profitably than we have indicated here. Nevertheless, our results clearly indicate that production of these commonly grown crops, when grown in only one season per year, alone cannot make most farming households in India prosperous. This is illustrated in Figure 5 where the annual crop profitability required to generate $2.15 per person per day is shown as a function of LPC, together with the percent distribution of LPC in India (from Figure 2). Farms with LPC values of 0.1 or less (39% of the total) would need to make annual profits of almost $8000 per hectare, and even those with LPC values of 0.8 have a profit requirement per hectare greater than the most profitable case in Table 1 (Mustard max, 2008–2011).

Figure 5.

Minimum crop net value per hectare per year (line) required to generate a personal daily income of $2.15 per person per day, as a function of land per capita. Bars represent the percentage distribution of LPC in Indian farm households.

Of course, 56% of Indian agricultural land is irrigated, the country has a cropping intensity of 156% (GOI, 2025) and so double- and even triple cropping are common. Where it is possible, sequential cropping increases annual gross output and profit per hectare, so the situation shown in Figure 5 would be improved. The implications of sequential cropping for smallholder farming households need to be considered and we do so in a subsequent paper.

Smallholder rural households do not confine themselves to the production of staple crops but commonly keep small and large livestock and operate home gardens. Although these clearly contribute to household income, dietary diversity, etc., any land-based enterprise is still subject to the limitations shown in Figure 5. It is also widely recognised that rural households do not base their livelihoods solely on farming (Rao and Charyulu, 2007; Sen, 2003) and may rely heavily on off-farm income. We suggest that in-depth studies of whole farms and whole households are required to understand the drivers of investment choice, the trade-offs that may be necessary, and how that shapes rural development.

Conclusions

Agricultural production is an intrinsically biological phenomenon in which inherently variable natural processes interact in complex ways with farm management and the environment but also with market factors to influence the pattern of production and so, ultimately, income (Pardey et al., 2014).

In this paper we have considered the temporal variation in profitability of individual crops and its consequences for the incomes of smallholder farmers in India. We have shown that, for the majority of farming households in India, low values of land per capita limit the absolute values of personal daily income that can be derived from crop production even if they were able to adopt best-practice technologies for any of the ten crops considered here.

Most farms are too small (and/or have to support too many household members) to generate even modest per capita incomes from crop production alone. Using even such a modest target CPDI of $2.15 – by no means a prosperous ambition – the large amounts of extra land and the correspondingly high additional variable costs required are unfeasibly high, beyond the means of most farm households, and carry unreasonable risks and uncertainties. Thus they represent a sub-optimal base for achieving widespread sustainable agricultural development in support of global goals of food security and environmental stewardship.

While it is widely known that crop production varies from season to season and from year to year, such variation is only apparent in retrospect and is impossible to predict in advance, that is, when choosing and planting a crop, and so there is a high degree of uncertainty concerning the likely returns on investment at the beginning of every cropping enterprise. Such uncertainty could be a major factor in the reluctance of smallholder households to adopt new technologies that may carry additional risks and so add to uncertainty.

Without additional income from local off-farm employment, from work on other people's farms, migration, remittances or businesses; smallholder farm households will struggle to cross ultra poverty lines even if they adopt best cropping practices. There is evidence however, that many, perhaps most, such rural households have income sources in addition to crop production (Birthal, 2025; Michler, 2020), including the rearing of small- and large livestock, which is widespread (although it should be noted that its income generation potential is dependent, directly or indirectly, on land area) and so generates similar low returns for smallholders (Attanasio and Augsburg, 2017; Gehrke and Grimm, 2018).

As we noted earlier, growing two- and three crops a year is common in India, particularly where supplementary irrigation is available. In theory, this should approximately double- or triple cropping income as it effectively doubles or triples values of land per capita per year. We consider the potential effect of multiple cropping on per capita incomes in a second paper in which we show that it alleviates the problem to some extent, but that growing additional crops requires additional investment that may be difficult for resource-poor households.

The limited potential for crop production to generate much income for households with small values of LPC, that is, most farmers in India, does not preclude the adoption of best practices designed to maximise yields (and minimise environmental degradation) but it does necessitate such households striking what they consider to be their best balance between pursuing on- and off-farm opportunities, and migration (Madhok et al., 2025). The details and drivers of such decision-making at individual- and household levels are not readily transparent as they are influenced by people's aspirations, ambitions and ever-changing circumstances and opportunities (Mausch and Harris, 2025).

This all poses a difficult challenge for policymakers who also have to strike a balance between supporting agriculture and other sectors of the economy. One obvious policy lever, given the preponderance of small-LPC farms in India, is land reform to enable the formation of larger farms as countries elsewhere in the world have done. However, land consolidation requires many people to give up farming and sell or rent out their land. Foster and Rosenzweig (2022) estimated that to maximise return per unit of land, the number of farms in India would have to be reduced by 82% and income per farm worker would rise by 68%. Without a corresponding increase in alternative employment that is unlikely to happen, although a coordinated multi-sectoral policy approach would, perhaps, increase the chance of success. However, the importance of agriculture for the economy and society is more than its contribution to income and poverty reduction (Chand, 2025). For example, farms are where people ‘come from’, call home and provide a refuge in times of trouble or hardship (Srivastava, 2020).

Small farms function as valuable components of an enduring social and economic system but require policy support that goes beyond own production to other facets of diversified rural livelihoods.

Supplemental Material

sj-docx-1-oag-10.1177_00307270261419516 - Supplemental material for Land per capita, year-to-year variability and uncertainty of return on investment affect household income from crop production: Evidence from India

Supplemental material, sj-docx-1-oag-10.1177_00307270261419516 for Land per capita, year-to-year variability and uncertainty of return on investment affect household income from crop production: Evidence from India by David Harris and Maxwell Mkondiwa in Outlook on Agriculture

Footnotes

Acknowledgements

Maxwell Mkondiwa acknowledges support from Sustainable Farming and Policy Innovations science programs of the CGIAR.

ORCID iDs

David Harris

Maxwell Mkondiwa

Funding

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

Declaration of conflicting interests

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

Supplemental material

Supplemental material for this article is available online.

References

Agarwal

Agrawal

(2017) Do farmers really like farming? Indian farmer in transition. Oxford Development Studies 45(4): 460–478.

Attanasio

Augsburg

(2017) Holy cows or cash cows? The economic return to livestock in rural India. Economic Development and Cultural Change 66(2). DOI: 10.1086/695138.

Binswanger HP (1986) Evaluating research system performance and targeting research in land-abundant areas of Sub-Saharan Africa. World Development 14(4): 469–475.

Birthal

(2025) Towards sustainable transformation of agri-systems in India: challenges and policy imperatives. Agricultural Economics Research Review 31(1): 7–17.

Cartsburg

Kaufmann

Nagel

(2024) Economic benefits through agroecological soil practices: evidence by ProSoil. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH. Available at: https://www.giz.de/en/worldwide/32181.html.

Chand

(2025) Reimagining agriculture for growth, poverty, nutrition, and sustainability. Agricultural Economics 56(3): 485–492.

Chand

Prasanna

Singh

(2011) Farm size and productivity: understanding the strengths of smallholders and improving their livelihoods. Economic & Political Weekly Supplement XLVI(26 & 27): 5–11.

Chand

Saxena

Rana

(2015) Estimates and analysis of farm income in India, 1983-84 to 2011-12. Economic & Political Weekly L(22): 139–145.

Consultative Group for International Agricultural Research (CGIAR) (2005) Agricultural research matters: achieving the millennium development goals, CGIAR, Washington, DC.

10.

Foster

Rosenzweig

(2022) Are there too many farms in the world? Labor market transaction costs, machine capacities, and optimal farm size. Journal of Political Economy 130(3): 636–680.

11.

Gathala

Laing

Tiwari

, et al. (2020) Enabling smallholder farmers to sustainably improve their food, energy and water nexus while achieving environmental and economic benefits. Renewable and Sustainable Energy Reviews 120: 109645.

12.

Gaurav

Mishra

(2015) Farm size and returns to cultivation in India: revisiting an old debate. Oxford Development Studies 43(2): 165–193.

13.

Gehrke

Grimm

(2018) Do cows have negative returns? The evidence revisited. Economic Development and Cultural Change 66(4): 673–707.

14.

Giller

Delaune

Silva

, et al. (2021) Small farms and development in sub-Saharan Africa: farming for food, for income or for lack of better options? Food Security 13(6): 1431–1454.

15.

Global Yield Gap Atlas (2025) Available at: http://www.yieldgap.org/ (accessed 19 August 2025).

16.

Godfray

Beddington

Crute

, et al. (2010) Food security: the challenge of feeding 9 billion people. Science 327: 812–818.

17.

Godfray

HCJ

Garnett

(2014) Food security and sustainable intensification. Philosophical Transactions of the Royal Society B: Biological Sciences 369: 20120273.

18.

Gollin

(2018) Farm size and productivity: lessons from recent literature. IFAD Research Series, no. 34. Available at: https://www.ifad.org/documents/d/new-ifad.org/research-series-34-pdf.

19.

Government of India (GOI) (2008) Manual on cost of cultivation surveys. Central Statistical Office, Ministry of Statistics and Programme Implementation, New Delhi. https://www.mospi.gov.in/sites/default/files/publication_reports/manual_cost_cultivation_surveys_23july08_0.pdf .

20.

Government of India (GOI) (2025) Annual report of the department of agriculture and farmers welfare. Ministry of Agriculture and Farmers Welfare, Government of India, New Delhi. https://agriwelfare.gov.in/Documents/AR_Eng_2024_25.pdf .

21.

Harris

Orr

(2014) Is rainfed agriculture really a pathway from poverty? Agricultural Systems 123: 84–96.

22.

Hayami

Ruttan

(1985) Agricultural Development: An International Perspective. Baltimore, USA: The Johns Hopkins University Press.

23.

Jat

Sapkota

Singh

, et al. (2014) Seven years of conservation agriculture in a rice–wheat rotation of Eastern Gangetic Plains of South Asia: yield trends and economic profitability. Field Crops Research 164: 199–210.

24.

Johnson

Singh

(2022) Trends in costs and incomes from crop production in India: analysis of costs of cultivation data, 2000-1 to 2019-20. Report prepared by the Foundation for Agrarian Studies for International Rice Research Institute. https://fas.org.in/wp-content/uploads/2024/02/Work-Package_2-new_FAS-IRRI-CCPC_Report.pdf .

25.

Kader

Jahangir

MMR

Islam

, et al. (2022) Long-term conservation agriculture increases nitrogen use efficiency by crops, land equivalent ratio and soil carbon stock in a subtropical rice-based cropping system. Field Crops Research 287: 108636.

26.

Kakraliya

Jat

Singh

, et al. (2018) Performance of portfolios of climate smart agriculture practices in a rice wheat system of western Indo-Gangetic plains. Agricultural Water Management 202: 122–133.

27.

Laik

Sharma

Idris

, et al. (2014) Integration of conservation agriculture with best management practices for improving system performance of the rice–wheat rotation in the Eastern Indo-Gangetic Plains of India. Agriculture, Ecosystems and Environment 195: 68–82.

28.

Lowder

Bhalla

Davis

(2025) Decreasing farm sizes and the viability of smallholder farmers: implications for resilient and inclusive rural transformation. Global Food Security 45: 100854.

29.

Lowder

Sánchez

Bertini

(2021) Which farms feed the world and has farmland become more concentrated? World Development 142: 105455.

30.

Lowder

Skoet

Raney

(2016) The number, size, and distribution of farms, smallholder farms, and family farms worldwide. World Development 87: 16–29.

31.

Lybbert

Shenoy

Bourdier

, et al. (2024) Striving to revive pulses in India with extension, input subsidies, and output price supports. American Journal of Agricultural Economics 106(3): 1167–1192.

32.

Madhok

Noack

Mobarak

, et al. (2025) Internal migration and the spatial reorganization of agriculture. NBER Working Paper No. 33854. National Bureau of Economic Research. https://www.nber.org/papers/w33854.

33.

Manjunatha

Anik

Speelman

, et al. (2013) Impact of land fragmentation, farm size, land ownership and crop diversity on profit and efficiency of irrigated farms in India. Land Use Policy 31: 397–405.

34.

Mausch

Harris

(2025) Aspirations as complementary entry points for rural development. In: Jones

Davies

(eds) ‘SenseMaker®: Praxis Makes Perfect. Distributed Ethnography in Theory and Practice. Colwyn Bay: Cognitive Edge Ltd., pp.199–207. ISBN 978-1-7353799-4-4. https://www.cifor-icraf.org/publications/pdf/chapters/KMausch_2025_Aspiration-as-complementary-entry.pdf.

35.

Michler

(2020) Agriculture in the process of development: a micro-perspective. World Development 129: 104888.

36.

National Sample Survey (2019) The situation assessment of agricultural households and land and holdings of households in Rural India, India. NSS 77th Round, Government of India, New Delhi.

37.

Nayak

Mkondiwa

Patra

, et al. (2025) Risk-return trade-offs in diversified cropping systems under conservation agriculture: evidence from a 14 year long-term field experiment in north-western India. European Journal of Agronomy 162: 127436.

38.

Nayak

Parihar

Mandal

, et al. (2022) Point placement of late vegetative stage nitrogen splits increase the productivity, N-use efficiency and profitability of tropical maize under decade long conservation agriculture. European Journal of Agronomy 133: 126417.

39.

Pal

Kumar

Patan

(2023) Investment planning to minimize climate risk in agricultural production: an optimization model for a semi-arid region in India. Climate Policy 23(4): 477–494.

40.

Pardey

Beddow

Hurley

, et al. (2014) A bounds analysis of world food futures: global agriculture through to 2050. Australian Journal of Agricultural and Resource Economics 58(4): 571–589.

41.

Rao

KPC

Charyulu

(2007) Changes in agriculture and village economies. Research Bulletin No. 21. International Crops Research Institute for the Semi-Arid Tropics, Patancheru.

42.

Reddy

Raju

Bose

(2020) Farmers’ income, indebtedness and agrarian distress in India. The Microfinance Review 12(1): 20–38.

43.

Rosegrant

Wong

Sulser

, et al. (2023) Benefit-cost analysis of increased funding for agricultural research and development in the global south. Journal of Benefit-Cost Analysis 14(S1): 181–205.

44.

Sanchez

Kamau

Grazioli

, et al. (2022) Financial profitability of diversified farming systems: a global meta-analysis. Ecological Economics 201: 107595.

45.

Sandhya

Singh

(2023) Study of productivity and economics of sunflower (Helianthus annuus L.) var. KBSH – 1 as influenced by nitrogen and spacing. International Journal of Environment and Climate Change 13(5): 439–443.

46.

Sen

(2003) Drivers of escape and descent: changing household fortunes in rural Bangladesh. World Development 31(3): 513–534.

47.

Sharma

Singh

(2024) Enhancing chickpea (Cicer arietinum L.) production through front-line demonstration in sub mountainous region of Punjab, India. Journal of Food Legumes 37(1): 95–100.

48.

Singh

Gaurav

Mishra

(2025) Dynamics of farm size-productivity relationship: evidence from semi-arid tropics of India, 1975-2014. Agricultural Economics 56(6): 874–894.

49.

Singh

(2021) Contribution of starter dose and seed inoculation in yield and economics of black gram under Hathras condition. Journal of Pharmacognosy and Phytochemistry 10(1): 158–162.

50.

Singh

Pandey

(2020) Economics and yield analysis of front-line demonstrations on sesame (Sesamum indicum L.). Journal of Rural and Agricultural Research 20(1): 96–98.

51.

Srivastava

(2020) COVID-19 and circular migration in India. Review of Agrarian Studies 10(1): 164–180.

52.

Tripathy

Meena

Dhar

, et al. (2023) Yield and economics of mustard + chickpea intercropping system under cropping systems and nutrient management practices in rainfed condition. Annals of Agricultural Research New Series 44(3): 318–324.

53.

United Nations (2015) Transforming our world: the 2030 agenda for sustainable development. Resolution adopted by the General Assembly.

54.

Van Ittersum

van Bussel

LGJ

Wolf

, et al. (2016) Can sub-Saharan Africa feed itself? Proceedings of the National Academy of Sciences 113(52): 14964–14969.

55.

Wei

Castro

Chhatre

, et al. (2025) Swapping rice for alternative cereals can reduce climate-induced production losses and increase farmer incomes in India. Nature Communications 16: 2108.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB