Overview

1. Household welfare and pricing of rice:Does the Large-Scale Field Model matter for Viet Nam?NGOC Q. PHAMand ANH H. LAVirtual Institute Seminar on Trade and PovertyGeneva, September 8-10, 2014

2. Overview • Introduction • The Large-Scale Field Model • Research objectives and questions • Related literature • Methodology and data • Findings • Policy recommendations

3. Viet Nam’s rice exports have soared since 2006

4. Does being the world’s biggest rice exporter matter for rice famer? • Even though one would expect higher rice exports to directly benefit rural households at all levels of well-being, most rice producers in Viet Nam are still poor, living on less than USD 2 per day. • The Vietnamese government has not implemented a number of price policy instruments seeking to ensure a minimum rate of return of 30 per cent for farmers. The main instruments have been “floor prices” for paddy both for ex- ports (minimum export free on board (FOB) price) and for purchases from rice farmers (minimum farm gate price for paddy).

5. Less than 5 per cent rice sales occur directly between farmers and exporters

6. What makes rice farming less profitable? • Existing Viet Nam’s rice value chain allows collectors to earn 10 times more than farmers (data in 2011 by Tran et al., 2013). • Farmers’ ability to bargain for higher prices is hampered by: • The market power of intermediaries: collectors usually set low prices. • Outstanding loans after harvest (accounted for more than 17 per cent of rice production cost). • The inability to store rice(lack of storage facilities and inadequate paddy drying technology). •  most farmers sell wet paddy to collectors

7. What prevents farmers from selling directly to exporters • 50 per cent of rice is exportedby public exporters through G2G contract, which do not requires high-quality of rice   SOEs lack incentive to improve performance  Vietnam’s export prices of rice fall. In order to to maximize their margin, public exporters prefer to buy rice from collectors rather than directly from farmers, as this allows them to lower paddy prices (avoid paying the official floor farm gate price for paddy). • Boat transportation and small fragmented fields (0.5 to 2 hectares) make high transportation and transaction costsfor exporters if they buy directly from farmers  collectors who own small boatsplay a key role in connecting small farmers and exporters.

8. What is the Large-Scale Field Model (LSFM)?

9. LSFM aims to increase price of paddy • By 9.9% in short-run:by reducing the role of intermediaries • By 11.4% in the longer-run: in case exporters agree to pass on 59 per cent (which equals the share of production costs in the export price of rice) of the increase in the export price of rice to farmers. Export price increase would be resulted from the higher and homogenous quality of paddy produced by farmers under the LSFM.

10. How the Large-Scale Field Model works? • The project also aims to consolidate land across farmers to reduce 14.1% of the cost of production through economies of scale.

11. How the Large-Scale Field Model works? • LSFM includes several measures that aim to improve farmers’ access to higher-quality inputs to subsequently increase 14% of average yields.

12. The research objective • Evaluation of the potential effects on the welfare of Vietnamese rice farmers of a pilot project that upgrades the rice export value chain (Large-Scale Field Model).

13. The research questions • What would be the effect of the implementation of the LSFM project on household welfare and poverty reduction? • Which of the main rice producing regions in Viet Nam – the Mekong River Delta or the Red River Delta – would be more suitable for this project? • Two complementary questions then concern the potential side effects of the LSFM: • Could food security be compromised if the LSFM results in higher prices for domestic production in foreign markets? • Would higher prices in one region be transmitted to other parts of the country through a price transmission mechanism?

14. Related literature • Households are affected by price changes: after a price increase, net consumers are worse off and net producers are better off. • Minot and Goletti (1998): rice export liberalization in Viet Nam would raise food prices but also increaseaverage real income and reduce poverty. • Benjamin and Brandt (2002): significant increases in the price of rice have a largely beneficial impact on rural household welfare. • Coxheadet al. (2012): negative effect of an increase in the price of rice on household welfare, especially among the poor.

15. Methodology • LSFM will affect household welfare through effects on: (a) farm gate prices of paddy, (b) productivity of rice farmers, and (c) production costs. • We model how these changes would affect household welfare, taking into account the ripple effect that a change in the farm gate price of paddy would have on other prices in the economy, and hence on household consumption, production, and wage income. • The methodology comprises three steps: • Input-output model to estimate the effects of paddy price on other products’ price; • Input-output model to estimate the effects on sectoral wages; • and first-order approximations (see Deaton, 1989) to estimate welfare impacts at the household level.

16. Methodology1.Estimation of price change • There are several ways of modelling the ripple effect a change in the market price of fertilizers and pesticides would have on other prices in the economy: • The econometric estimation model (Nicitaet al., 2005; Balatet al., 2009) and the global simulation model(Francois and Hall, 2009):data limitation; concordance of trade data classification with the living survey classification. • Computable General Equilibrium (CGE)model (Chen and Ravallion, 2004): limitation of time series data required to estimate the elasticities and/or data required to calibrate the parameters of the model. • The cost-push Leontief price model(Miyazawa, 1976; Oosterhaven, 1996; Dietzenbacher, 1997; ten Raa, 2005; and Miller and Blair, 2009)

17. Methodology1.Estimation of price change (cont’) (1) To estimate the ripple effect a change in the prices of fertilizers and pesticides would have on other prices in the economy, split A into: (2) • The cost-push Leontief price model: The cost-push Leontief price model becomes: (3) The price changes in other S sectors due to price changes in P sectors: (4)

18. Methodology2. Estimation the effects on sectoral wages (5) Take pp and vsas exogenous variables, whereas psand vpare endogenous variables, under the assumption that capital stock coefficients are constant in the short term and thus have not been affected by price changes, we obtain: • Equation (3) can be modified as: (6)

19. Methodology3. Estimation of welfare effects • Study the distributional effects of price changes in relation to household characteristics, particularly living standards and geographical locations(Deaton, 1989). • For each household, the welfare impactis: (7)

20. Data sources • Data on monthly domestic prices of paddy and rice are available from the Information Center for Agriculture and Rural Development for the period of January 2008 to December 2013. • Estimation of the price changes: the national input-output table of 2007 published by the General Statistics Office of Vietnam. • Vietnam Household Living Standards Survey 2010 to estimate household welfare effects.

21. Estimation of the top 20 price increases

22. Scenarios

23. Baseline – why not the Red River Delta?

24. Distribution of income from rice production

25. Farm size

26. Rice income by farm size

27. Total welfare effects – short and long run (1) When the poverty line is defined as the income of the richest among the 10 per cent poorest households in rural areas of the Mekong River Delta, the poverty rate is 5.5 per cent. (2) When the poverty line is defined as the income of the richest among the 20 per cent poorest households in rural areas of the Mekong River Delta, the poverty rate is 13.46 per cent.

28. Total welfare effects– Scenario 1

29. Total welfare effects– Scenario 4

30. Side effects of LSFM • In 2030, the expected output would be far in excess of national food security needs even under the worst-case scenario(Jaffeeet al., 2012) • Econometric estimations (use panel data on weekly domestic paddy prices from 1stJanuary 2008 to 20th August 2013): • The estimated pass-through elasticity is close to zero but still positive and statistically significant an increase in the farm gate price of paddy following the LSFM would only marginally affect the domestic price of rice in the rest of Viet Nam.

32. Policy recommendation • The results of our analysis show that farmers from the Mekong River Delta should be given priority for application of the LSFM. • The estimations of the welfare and poverty reduction effects of the LSFM in the Mekong River Delta also suggest that the LSFM might be a better policy option than setting floors for export prices and farm gate prices of paddy. • The effects of the LSFM would not only improve household welfare in the region but also expected to foster Vietnamese rice exports. As state-owned exporters may have fewer incentives to implement the changes proposed by the LSFM project, private exporters would likely be better candidates to lead its implementation. Successful implementation of the LSFM would however require some level of competition among Vietnamese rice exporters and hence access to the rice export quota.