Abstract
Selected Issues
Agriculture: Key Issues and Reforms1
This chapter takes stock of key issues in and identifies important reforms for India’s agricultural sector. It highlights long-term structural bottlenecks, including low agricultural productivity, large distortions particularly those induced by government interventions, and marketing issues. Sustained inclusive growth requires agricultural sector reforms, which should focus on reducing supply-side constraints, building more integrated markets, boosting productivity, and addressing market distortions.
1. Agriculture is the backbone of the Indian economy. The agricultural sector contributes around a fifth of GDP and provides employment for about half of the labor force, and farm income is a key determinant of rural consumption. Given the sheer size of food’s weight (almost half) in the consumer price basket, changes in agricultural prices have a major impact on consumer prices.
2. India’s economic transformation has progressed steadily. Country experiences suggest economic development is associated with the transition from agriculture to higher value-added sectors such as manufacturing or services. Over time, the share of agricultural sector declines and the number of agricultural workers falls. Nevertheless, such modernization must occur alongside increased agricultural productivity growth, while ensuring adequate farm incomes and food security (Subramanian (2018)).
Share of Agriculture and GDP Per Capita
(1960, 1980, 2000, 2016)
Citation: IMF Staff Country Reports 2018, 255; 10.5089/9781484373200.002.A005
Sources: CDM Next, World BankShare of Agriculture and GDP Per Capita
(1960, 1980, 2000, 2016)
Citation: IMF Staff Country Reports 2018, 255; 10.5089/9781484373200.002.A005
Sources: CDM Next, World BankShare of Agriculture and GDP Per Capita
(1960, 1980, 2000, 2016)
Citation: IMF Staff Country Reports 2018, 255; 10.5089/9781484373200.002.A005
Sources: CDM Next, World Bank3. Agricultural sector policy reforms continue to be a priority. The Government of India (GOI) has called for doubling farmers’ incomes by 2022 as one of its top priorities. Recent agricultural policy initiatives are crucial to reducing production risk and improving the competitiveness of agricultural markets. These initiatives are, for instance, the assured irrigation initiative under the Pradhan Mantri Krishi Sinchayee Yojana launched in July 2015, the comprehensive crop insurance scheme Pradhan Mantri Fasal Bima Yojana launched in February 2016, the common electronic trading platform for a National Agriculture Market (e-NAM) launched in April 2016, and Gramin Agricultural Markets (GrAMs) launched in 2018, as well as the adoption of Model Agricultural Produce and Livestock Marketing Act (Model Act) of 2017 to facilitate private sector investment in the agricultural sector and make agricultural marketing and distribution more flexible.
4. This study attempts to take stock of key issues and identify reform areas. The chapter highlights long-term structural bottlenecks in the agricultural sector, including low agricultural productivity, large distortions particularly those induced by government interventions, and marketing issues. It empirically analyzes the determinants of agricultural production and yield across Indian states and examines the role of structural measures, the effectiveness of government spending on agricultural sector, and the implications of price intervention. Based on the findings, it discusses policy recommendations.
5. Long-standing issues are related to low agricultural productivity. India’s agricultural labor productivity is less than a third of that of China and only about one percent of that of the frontier—the United States (Subramanian (2018)). Insufficient agricultural infrastructure including irrigation systems and cold storage leads to significant production risk, waste, and losses to farmers. Fractured land holding makes it difficult to gain economies of scale. Land-leasing regulations remain tight, hindering an expansion into larger-scale agricultural business.
Overall Agricultural Productivity
(Gross Value Added per worker, USD, 2005 prices)
Citation: IMF Staff Country Reports 2018, 255; 10.5089/9781484373200.002.A005
Source: “Transforminglndiaan Agriculture: By Loving Some AgricultureLess and the Rest More”, ArvindOverall Agricultural Productivity
(Gross Value Added per worker, USD, 2005 prices)
Citation: IMF Staff Country Reports 2018, 255; 10.5089/9781484373200.002.A005
Source: “Transforminglndiaan Agriculture: By Loving Some AgricultureLess and the Rest More”, ArvindOverall Agricultural Productivity
(Gross Value Added per worker, USD, 2005 prices)
Citation: IMF Staff Country Reports 2018, 255; 10.5089/9781484373200.002.A005
Source: “Transforminglndiaan Agriculture: By Loving Some AgricultureLess and the Rest More”, Arvind6. The old agricultural support framework leads to large distortions. It involves three main policy interventions—input subsidies, minimum support prices (MSPs), and the Public Distribution System (PDS).2 This framework has created significant distortions; however, political obstacles deterred efforts to move forward with reforms (Fan et al, 2007). The issues are, for instance:
Agricultural subsidies continue to weigh on the government budget, crowd out productive spending, and be often poorly administered, although significant progress has been made to streamline various subsidies in recent years.
Past MSP hikes for rice and wheat, combined with the government’s massive cereal stockpiling, resulted in production distortions, sharp swings in stocks, and episodes of high food inflation (IMF, 2017). The MSP has been implemented to ensure remunerative prices to farmers, but paradoxically, as farmers may not be aware of the MSP, its benefits may not reach them (Chatterjee and Kapur (2016)). Historically, substantial increases in the MSP were generally followed by rising inflation in key crops, fueling inflationary pressures. The FY2018/19 Budget announced that the MSP has been declared for all Rabi crops at least 1.5 times of the cost of production and the MSP for the unannounced Kharif crops will also be fixed in a similar manner. Nevertheless, the implementation remains unclear.
Weaknesses in the PDS manifest large leakages and operating inefficiencies (IMF, 2016). Significant leakages—subsidized grains not reaching poor households—are estimated from 40 to 60 percent and may be much higher in some states. The operating costs of the Food Corporation of India (FCI)—the central government’s entity responsible for procurement, storage, transportation, and bulk allocation of food grains to the State Governments—are high, with the FCI’s costs of acquiring, storing, and distributing food grains approximately 40 to 50 percent more than the procurement prices.
7. Problems in agricultural marketing are being addressed. Marketing is governed by the Essential Commodities Act and the state-level Agricultural Produce Marketing Committee (APMC) Acts, which empower the central and state governments to regulate and control production, distribution, marketing, and pricing of commodities identified as essential for consumers. The intention of the APMC Acts was to ensure that farmers were offered fair prices in a transparent manner. Nevertheless, this has turned rural agricultural markets (mandis) into local monopsonies by restricting free entry, causing lack of competition and transparency, and discouraging investments by the private sector, resulting severe governance challenges (Chatterjee and Kapur (2016)). While empirical evidence on agricultural price convergence suggest retail prices converge across India, price variation at APMC mandis across India persists over time.3 To date, several initiatives have taken off to address problems in agricultural marketing, although it may take some time to resolve some implementation issues including to fully automate and improve internet infrastructure, as well as increase training and capacity (Aggarwal et al, (2017)). The 2017 Model Act should also help improve efficiency and transparency.
Empirical Analysis
8. The analysis estimates an agricultural production function as in Lin (1992), using cross-state data (Tables 1-2). The main inputs are land, labor, capital as proxied by credit to the agriculture sector, and fertilizer. Various policy variables are included to examine their impacts on production.
Production (Yit) is measured as the production of major crops including food grain, rice, pulses, sugarcane, cereals, and wheat weighted by the share of area cropped, with i being an index for states. Xit is a K-dimensional policy vector which includes the government’s capital expenditure, infrastructure spending, and MSP, among others, with coefficients indexed k = 1, …, K. Data are annual spanning from 1980 to 2016 and covering 23 states, from Centre for Monitoring Indian Economy (CMIE) and CEIC database. The panel is unbalanced. The model was also estimated for the weighted yield, replacing weighted production. Conclusions are broadly unchanged.4
9. Infrastructure spending is found to be strongly associated with the improvement of agricultural production and yield. Labor, land, fertilizer, and credit explain nearly 70 percent of the variation in production. Broad infrastructure development, as proxied by capital expenditure, appears to be positively correlated with production. Specific efforts at structural improvements in agriculture, for example improving food storage and irrigation, are also significant and help improve explanatory power. Spending on food storage impacts production with the same magnitude as credit. Land that is irrigated is associated with an increase of production by about 25 percent. Higher MSPs contribute to higher quantities produced. This impact comes with market distortions as highlighted previously and could add to inflation and fiscal costs.5
Policy Recommendations
10. Sustained inclusive growth requires agricultural sector reforms. In line with recommendations by NITI Aayog and in the Economic Survey 2018, the strategy should focus on raising productivity, reducing production risk, and fostering more competitive agricultural markets. This study highlights the importance of reducing supply-side constraints, building more integrated markets, boosting productivity, and addressing market distortions. Possible policy actions include:
Recent policy initiatives such as the assured irrigation system, the introduction of e-NAM, and the development of GrAM are welcome and, despite gradual implementation, promise to reduce production risk, increase competitiveness, and improve transparency in state markets.
To further reduce vulnerability, there is also the need to continue to address long-term structural bottlenecks, including in irrigation and other infrastructure.
Boosting agricultural productivity requires more efficient use of inputs, improved agricultural technology, research and development, and education.
As MSPs could skew farmers’ production decisions, add to inflation, and enlarge the fiscal burden, their use (backed by assured procurement) should only be temporary and limited to correcting market failures.
To address distortions, more needs to be done to revamp government procurement processes and the PDS, including to restructure the role of the FCI via outsourcing of cereal procurement and stocking operations and check leakages in the PDS.
Various agricultural subsidies are being streamlined especially through direct benefit transfers, and should be further reduced going forward.
India: Regression Results for Agricultural Production and Yield 1/
All variables are in logs. The dependent variable is the weighted average production of food grains, rice, pulses, sugarcane, cereals, and wheat by share of area sown in columns (1)-(3), the weighted average production of rice, pulses, cereals, and wheat by share of area sown in columns (4)-(7), the weighted average yield of food grains, rice, pulses, sugarcane, cereals, and wheat by share of area sown in columns (8)-(10). Labor is the sum of agricultural laborers and cultivators in rural areas. Land is the sum of the area sown for the crops in the dependent variable. Fertilizer is total consumption of fertilizer in kg. Agricultural credit is the total outstanding agricultural credit in all scheduled commercial banks. Capex development and revenue expenditure on water and food storage are nominal values from state government finances. MSP is the weighted average by area sown for the crops considered. Net irrigation is the total area irrigated in hectares. Rainfall deviation is the percent deviation from average rainfall. All columns include state fixed effects. Standard errors are in parentheses. * p<0.05; ** p<0.01
India: Regression Results for Agricultural Production and Yield 1/
| Dependent variable: weighted production | Dependent variable: weighted yield | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| All products | of which: MSP products | All products | ||||||||
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | |
| Labor | 0.29 | 0.33 | 0.12 | 0.03 | 0.43 | 0.35 | 0.10 | 0.50 | 0.55 | 0.32 |
| (3.56)** | (3.27)** | (1.43) | (0.29) | (2.75)** | (2.33)* | (0.71) | (6.28)** | (5.44)** | (3.53)** | |
| Land | 0.86 | 0.93 | 0.82 | 0.65 | 1.71 | 1.74 | 1.71 | |||
| (11.49)** | (11.75)** | (11.58)** | (7.82)** | (13.33)** | (13.31)** | (12.58)** | ||||
| Fertilizer | 0.13 | 0.13 | 0.22 | 0.13 | −0.03 | −0.04 | 0.00 | 0.07 | 0.08 | 0.16 |
| (5.44)** | (4.29)** | (6.15)** | (4.73)** | (0.62) | (0.80) | (0.07) | (2.94)** | (2.56)* | (4.36)** | |
| Agriculture Credit | 0.12 | 0.04 | 0.05 | 0.12 | 0.04 | 0.04 | 0.01 | 0.14 | 0.05 | 0.07 |
| (14.57)** | (2.32)* | (5.37)** | (11.75)** | (1.65) | (1.33) | (0.22) | (16.10)** | (2.84)** | (6.17)** | |
| Capex Development | 0.09 | 0.01 | 0.08 | |||||||
| (4.64)** | (0.60) | (4.02)** | ||||||||
| MSP | 0.15 | 0.13 | 0.20 | |||||||
| (2.44)* | (2.05)* | (3.93)** | ||||||||
| Revenue Expenditure Water | 0.00 | −0.04 | 0.01 | |||||||
| (0.27) | (2.14)* | (0.63) | ||||||||
| Revenue Expenditure Food Storage | 0.04 | 0.03 | 0.03 | |||||||
| (5.51)** | (2.58)* | (4.60)** | ||||||||
| Net Irrigation | 0.25 | 0.23 | 0.24 | |||||||
| (4.58)** | (3.19)** | (4.05)** | ||||||||
| Rainfall Deviation | −0.01 | −0.01 | −0.01 | |||||||
| (2.32)* | (1.93) | (1.79) | ||||||||
| Constant | 0.35 | −1.13 | 1.05 | 6.26 | −6.68 | −5.49 | −3.82 | −2.21 | −3.29 | −1.73 |
| (0.26) | (0.69) | (0.82) | (4.04)** | (2.63)** | (2.20)* | (1.66) | (1.77) | (2.03)* | (1.35) | |
| Adjusted R2 | 0.68 | 0.70 | 0.72 | 0.58 | 0.57 | 0.61 | 0.69 | 0.67 | 0.67 | 0.67 |
| Number of observations | 705 | 482 | 409 | 723 | 325 | 254 | 237 | 705 | 482 | 409 |
All variables are in logs. The dependent variable is the weighted average production of food grains, rice, pulses, sugarcane, cereals, and wheat by share of area sown in columns (1)-(3), the weighted average production of rice, pulses, cereals, and wheat by share of area sown in columns (4)-(7), the weighted average yield of food grains, rice, pulses, sugarcane, cereals, and wheat by share of area sown in columns (8)-(10). Labor is the sum of agricultural laborers and cultivators in rural areas. Land is the sum of the area sown for the crops in the dependent variable. Fertilizer is total consumption of fertilizer in kg. Agricultural credit is the total outstanding agricultural credit in all scheduled commercial banks. Capex development and revenue expenditure on water and food storage are nominal values from state government finances. MSP is the weighted average by area sown for the crops considered. Net irrigation is the total area irrigated in hectares. Rainfall deviation is the percent deviation from average rainfall. All columns include state fixed effects. Standard errors are in parentheses. * p<0.05; ** p<0.01
India: Regression Results for Agricultural Production and Yield 1/
| Dependent variable: weighted production | Dependent variable: weighted yield | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| All products | of which: MSP products | All products | ||||||||
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | |
| Labor | 0.29 | 0.33 | 0.12 | 0.03 | 0.43 | 0.35 | 0.10 | 0.50 | 0.55 | 0.32 |
| (3.56)** | (3.27)** | (1.43) | (0.29) | (2.75)** | (2.33)* | (0.71) | (6.28)** | (5.44)** | (3.53)** | |
| Land | 0.86 | 0.93 | 0.82 | 0.65 | 1.71 | 1.74 | 1.71 | |||
| (11.49)** | (11.75)** | (11.58)** | (7.82)** | (13.33)** | (13.31)** | (12.58)** | ||||
| Fertilizer | 0.13 | 0.13 | 0.22 | 0.13 | −0.03 | −0.04 | 0.00 | 0.07 | 0.08 | 0.16 |
| (5.44)** | (4.29)** | (6.15)** | (4.73)** | (0.62) | (0.80) | (0.07) | (2.94)** | (2.56)* | (4.36)** | |
| Agriculture Credit | 0.12 | 0.04 | 0.05 | 0.12 | 0.04 | 0.04 | 0.01 | 0.14 | 0.05 | 0.07 |
| (14.57)** | (2.32)* | (5.37)** | (11.75)** | (1.65) | (1.33) | (0.22) | (16.10)** | (2.84)** | (6.17)** | |
| Capex Development | 0.09 | 0.01 | 0.08 | |||||||
| (4.64)** | (0.60) | (4.02)** | ||||||||
| MSP | 0.15 | 0.13 | 0.20 | |||||||
| (2.44)* | (2.05)* | (3.93)** | ||||||||
| Revenue Expenditure Water | 0.00 | −0.04 | 0.01 | |||||||
| (0.27) | (2.14)* | (0.63) | ||||||||
| Revenue Expenditure Food Storage | 0.04 | 0.03 | 0.03 | |||||||
| (5.51)** | (2.58)* | (4.60)** | ||||||||
| Net Irrigation | 0.25 | 0.23 | 0.24 | |||||||
| (4.58)** | (3.19)** | (4.05)** | ||||||||
| Rainfall Deviation | −0.01 | −0.01 | −0.01 | |||||||
| (2.32)* | (1.93) | (1.79) | ||||||||
| Constant | 0.35 | −1.13 | 1.05 | 6.26 | −6.68 | −5.49 | −3.82 | −2.21 | −3.29 | −1.73 |
| (0.26) | (0.69) | (0.82) | (4.04)** | (2.63)** | (2.20)* | (1.66) | (1.77) | (2.03)* | (1.35) | |
| Adjusted R2 | 0.68 | 0.70 | 0.72 | 0.58 | 0.57 | 0.61 | 0.69 | 0.67 | 0.67 | 0.67 |
| Number of observations | 705 | 482 | 409 | 723 | 325 | 254 | 237 | 705 | 482 | 409 |
All variables are in logs. The dependent variable is the weighted average production of food grains, rice, pulses, sugarcane, cereals, and wheat by share of area sown in columns (1)-(3), the weighted average production of rice, pulses, cereals, and wheat by share of area sown in columns (4)-(7), the weighted average yield of food grains, rice, pulses, sugarcane, cereals, and wheat by share of area sown in columns (8)-(10). Labor is the sum of agricultural laborers and cultivators in rural areas. Land is the sum of the area sown for the crops in the dependent variable. Fertilizer is total consumption of fertilizer in kg. Agricultural credit is the total outstanding agricultural credit in all scheduled commercial banks. Capex development and revenue expenditure on water and food storage are nominal values from state government finances. MSP is the weighted average by area sown for the crops considered. Net irrigation is the total area irrigated in hectares. Rainfall deviation is the percent deviation from average rainfall. All columns include state fixed effects. Standard errors are in parentheses. * p<0.05; ** p<0.01
References
Aggarwal, N., S. Jain, and S. Narayanan, 2017, “The Long Road to Transformation of Agricultural Markets in India: Lessons from Karnataka.” Economic & Political Weekly, Vol LII, No.41, pp 47–55.
Chatterjee, S. and D. Kapur, 2016, “Understanding Price Variation in Agricultural Commodities in India: MSP, Government Procurement, and Agriculture Markets.” Presented at India Policy Forum July 12-13, 2016.
Fan, S., A. Gulati, and S. Dalafi, 2007, “Reforms and Development in China and India.” In Gulati, A. and Fan, S. (eds), The Dragon and the Elephant: Agricultural and Rural Reforms in China and India. Baltimore: Johns Hopkins University Press.
The Government of India, 2018, “Climate, Climate Change, and Agriculture.” In Economic Survey 2018.
International Monetary Fund, 2016, “India: Reorienting the Role of the Food Corporation of India (FCI).” IMF Country Report No. 16/75 (International Monetary Fund: Washington).
International Monetary Fund, 2017, “India: 2017 Article IV Consultation—Press Release; Staff Report; and Statement by the Executive Director for India,” IMF Country Report No. 17/54 (International Monetary Fund: Washington).
International Monetary Fund, 2018, “Domestic Market Integration and the Law of Once Price in India.” Staff Note.
Lin, J. Y., 1992, “Rural Reforms and Agricultural Growth in China.” The American Economic Review, Vol. 82, No. 1, pp 34–51.
Subramanian, A., 2018, “Transforming Indian Agriculture: By Loving Some Agriculture Less and the Rest More.” Lecture series at Indian Institute of Technology Delhi.
Prepared by Racha Moussa and Piyaporn Sodsriwiboon.
The PDS has evolved as a system for management of scarcity and for distribution of food grains at affordable prices. It largely governs the procurement of India’s main agricultural commodities namely wheat, rice, sugar, and pulses, among others, and distributes to targeted poor households.
IMF (2018) applies panel unit root tests to examine whether retail prices have converged over time. It uses monthly data for 15 crops by city from 2010 to 2016. Preliminary results suggest that the law of one price holds for various crops. On the other hand, Chatterjee and Kapur (2016) analyze the spatial variation in wholesale prices of the principal cereal crops (rice and wheat) in all APMC mandis across India and within each state. It finds spatial variations in real prices of agricultural commodities are large and persist through time.
The model estimated without land retained a high explanatory power.
Regressions that include MSP cover rice, pulses, cereal, and wheat.
