FFTC Agricultural Policy Platform (FFTC-AP)

ABSTRACT

Under the requirements of the AMS reduction and competitiveness improvement in the rice sector, the government has been implementing the Rice Production and Marketing Areas (RPMA) program in Taiwan since 2005. It rewarded operation units with price that is higher than the guaranteed price of government through making contact with rice farmers, and to run a business model by integrated value chain. The purposes of this study are to evaluate the policy performance in terms of government expenditure and farmers’ income, and estimate profit efficiency of operation unit in the RPMA program. It found that the program of the government’s purchasing paddy reduction and expenditure savings has some effects. Farmers’ income would be extra increased from participating in the program. Besides, profit efficiency of operation units which participated in the program is also higher compared to than non-RPMA. It revealed that the core strategy for the RPMA program is to enhance rice value-added.

Keywords: rice, price support, profit efficiency, metafrontier model

INTRODUCTION

Rice is the most important industry in the agricultural sector in Taiwan, and it is also an industry that invests the most resources and government expenditures. However, thr government purchases rice at a guaranteed price that distorts market mechanism, reduces efficiency of resource allocation, and causes a heavy financial burden on the government. At the same time, it also faces pressure from the WTO to reduce AMS. Therefore, the government started to implement the "Rice Production and Marketing Area"(RPMA) program in 2005, hoping to reduce the government's purchase expenditure and AMS through reward payment for grain merchants to encourage making production contract with rice farmers. The RPMA program is accompanied by the common production operations and marketing activities and it can ensure the quality of domestic rice and establish its own brand. It is expected to benefit to the rice industry in terms ofupgrading quality standards and competiveness.

The purpose of this paper is to evaluate the RPMA program performance in terms of government expenditure, farmer’ revenue and profit efficiency of grain merchants, so as to provide a reference basis for rice industry development and policy reforms in Taiwan. The rest of the paper is organized as follows. The next section briefly describes the features and current situation of RPMA. Section 3 evaluates RPMA by rice policy review and its influence on government’s purchasing quantity and expenditures and farmers’ revenue. Section 4, on the other hand, focuses on profit efficiency estimation and its determinations for grain merchants who participated in the RPMA program. Finally, the last section is the conclusion and outlook for the RPMA program.

I. Rice Production and Marketing Area (RPMA)

Features

In order to avoid the impact from importing rice in the domestic rice industry Taiwan has the joined World Trade Organization in 2002 and emphasized domestic rice production with its own brand that symbolizes good quality. The government has been trying to do rice industry adjustment since 2005. The promotion of the RPMA program is one of the adjustment strategies. Government encourages Farmers’ Associations and grain merchants as operation unit to coordinate with local rice farmers, nursery operators and millers to establish a RPMA which at least 50 hectares. It is expected to have advantages in in terms of the RPMA’s production integration and economic scale.

The most important thing is that operation units have to make contract with participating farmers at higher price compared to the guaranteed price of government purchases. It could raise farmers’ income and reduce government purchasing rice and expenditure. On the other hand, operation units can get reward payment from the government to implement the RPMA program to support higher price under contract and related spending on production integration, farmers’ education and product marketing.

Although there are multiple meanings for rice farmers, the government, grain merchants and even consumers, the final target of the RPMA program is to enhance competitiveness and value-added for rice industry in Taiwan.

Implementation

The RPMA program has been in operation since 2005. It started with 27 operation units which included Farmers’ Associations and grain merchants with 4,908 hectares total, then increasing to 63 operation units with a total of 23,292 hectares as shown in Table 1. Because of affording payment at higher contract price to participating farmers, the number of operation units failed to increase as expected. It still less than 7% in terms of total rice plant hectares. Fortunately, the number of operation units and participating hectares had significantly increased in the past two years.

Table 1. Implementation of the RPMA program during 2005-2017

Source: Agriculture and Food Agency, Council of Agriculture.

Basically, the reward payment provided by the government is a key factor to attract Farmers’ Associations and grain merchants to participate in the RPMA program. The correlation coefficient is 0.86 between reward payment and participated hectares. The government could have a balance consideration on the reward payment and purchasing expenditure saving to decide whether to increase reward payment for expansion and exclusion of the participated hectares.

The operation units could be Farmers’ Association and grain merchant with different business culture and objectives. There were 21 Farmers’ Associations and grain merchants and 42 grain merchants who participated in the RPMA program in 2017. Under the consideration of market size and capital stock, Farmers’ Association and grain merchants sell milled rice mainly and partly from what it purchased paddy rice from the RPMA, respectively.

Table 2 indicates that average contract price of Farmers’ Association is 27.13 NTD/Kg, which is higher than grain merchants, 25.28 NTD/Kg. It reveals that Farmers’ Association has been supporting farmers’ income instead of just searching for profit maximization. The average participated area for Farmers’ Association is 289 hectares, which is smaller than grain merchants, 413 hectares. It would affect benefit of economic scale. The average scale for each participated farmers’ of Farmers’ Association is 1.62 hectares, which is lower than grain merchants, 1.86 hectares, as well.

Table 2. Current situation of the RPMA program in 2017

Source: Agriculture and Food Agency, Council of Agriculture.

The RPMA program is a kind of contract farming which is based on mutual benefit for farmers and operation units participation. It has been in existence for many years as a means of organizing the commercial agricultural production of both large-scale and small-scale farmers. Eaton and Shephered (2001) pointed out contract farming is beneficial to farmers’ price risk that is often reduced as many contracts specify prices in advance and can open up new markets which would otherwise be unavailable to small farmers. Operation units also gained more consistent quality and quantity which can be obtained than if purchases were made in the open market. There are more than 70% of operation units using brand that is endorsed by the RPMA program to increase rice price and competitiveness (Chen and Yang, 2007).

Since there are different production condition in Taiwan, there is only one crop season in the Northern part that is different from other parts with two crop seasons. Western farmers have been searching for quantity increase, unlike eastern farmers who emphasize quality improvement. Therefore, for regional distribution of participating in the RPMA program, the East part participate most, 58%, and contract price, 25.49 NT/Kg, which is also higher than other parts.

II. Policy evaluation

Rice industry and policy

Rice is staple food and the most important crop in Taiwan. Harvested area is 273,837 hectares, where first crop is 168,872 hectares and second crop is 104,965 hectares in 2017. Production quantity in terms of paddy rice is 1,587,776 MT.

For ensuring food security and raising farmers’ income, government has been implementing policy of purchasing paddy rice at guaranteed price which is higher than market price. The purchasing system is made up of three layers: planned purchase, guidance purchase and surplus purchase with different guaranteed price and quantities. The weighted average guaranteed price is 24.91 NT/Kg, which is higher than farm gate price 23.91 NT/Kg, as shown in Table 3. Government purchased paddy rice for 312,033 MT, which accounted for nearly 20% of total production in 2017.

Theoretically, quantity purchased by government depends on guaranteed price and price difference between guaranteed price and farm gate price. In the past, government raised guaranteed price several times to raise farmers’ income. It also stimulated rice plant area and production expansion which resulted in the problem of much purchasing quantity and excess inventory. In 2011, government raised guaranteed price by 3 NT/KG. It brought about significant jumps in purchasing quantity 406,223 MT and expenditure over 10 billion NTD in 2013.

Although there were many contributions on food security and farmers’ income, there have also been many criticisms always on the policy of purchasing paddy rice at guaranteed price, including market mechanism distortion, government financial burden, resources allocation inefficiency and crops production imbalance, etc. However, due to political consideration, it is hard to terminate the policy. The reason for promoting the RPMA program is trying to partially substitute purchasing paddy rice at guaranteed price and solve some related problems.

Table 3.  Government paddy purchase price and quantity

Source: Agriculture and Food Agency, Council of Agriculture (2017), Taiwan Food Statistics Book.

Policy evaluation on the RPMA program

Since emphasizing on quantity increase rather than quality improvement under oriented by the policy of purchasing paddy rice at guaranteed price in Taiwan, the government continually promotes a series of rice industry adjustments, which includes the RPMA program. Based on experiences of farmers who participated in the planned purchase and guidance purchase, the estimated reduction of the government’s purchasing paddy rice was 46,584 MT with expenditure saving 1.1 billion in 2017 as shown in Table 4. Compared with reward payment 0.12 billion. The government benefitted from a net expenditure savings of about 1 billion NTD.

Besides, farmers’ income would be extra increased from contract price, which is higher than government’s guaranteed price, by at least 0.1billin total. For instance, farmers’ income increased by 0.145 billion NTD total in 2017; in other words, each farmer’s income increased by 11,178 NTD average, or each hectare increased by 6,227 NTD average.

Table 4.  Policy benefit of the RPMA program

OPERATION PERFORAMNCE EVALUATION

Considerations of operation units

Participants of the RPMA program are government, farmers, operation units (i.e. Farmers’ Association and grain merchants). In the above analysis, the program is beneficial to government’s expenditure saving and raising of farmers’ income. However, the key factor is operation unit participation. If operation units would improve their efficiency and profit, it is expected that more operation units would participate in the RPMA program in the future.

Contrast to production efficiency estimation used inputs and out data, this paper emphasizes profit efficiency estimation used cost and revenue data, which is consistent with operation units consideration about profit objective.

O’Donnell, Battese and Rao (2008) suggested application of Metafrontier model to estimate profit efficiency of operation units with different scale, area and business culture.

Model specification

Assume the form of profit function of operation units is random translog. We can build a profit Metafrontier function as follow (Battese et al., 2004; O’Donnell et al., 2008):

Where πis operation unit’s standardized profit deflated by milled rice sale price, W₁ , W₂ and W₃ are standardized paddy rice purchase price, process cost and salary, respectively. βare parameters. Error term consists of random error (v_it(j)) and profit inefficiency (u_it(j)), the former assumed to be normal distribution and the latter is actual profit for the i operation unit in t year which might deviate from optimal profit in the group j; i.e, it means profit inefficiency. It is assumed that v_it(j) is independent of u_it(j).

We will estimate profit efficiency step by step. Firstly, application of Stochastic Frontier Analysis (SFA) and Maximum Likelihood Estimation (MLE) to estimate profit efficiency of the i operation unit in group j. Secondly, constitute random profit function of each group as follow:

Where i=1,2,… N_j, t=1,2,…T, and j=1,2, …K represent different operation unit, time and group, respectively.

The purposes of random profit function are to estimate profit efficiency of the RPMA or not and its profit efficiency ratio. Thirdly, we then judge estimated profit efficiency difference by Likelihood ratio. If reject null hypothesis, it means there is different profit efficiency in participation of the RPMA program or not. Finally, application of GAUSS software suggested by Battese et al. (2004) incorporate estimated parameters of each group into linear programming (LP) and quadratic programming (QP) to estimate optimal parameter of profit Metafrontier as follow:

Where π_it is profit on Metafrontier. It represents potential profit for operation unit with opitimal resource allocation and as a basis for Metaprofit efficiency and  Metaprofit ratio calculation.

Empirical results

Profit efficiency

Estimated parameters of operation units in the PRAM, non-PRAM and total are shown in appendix table A1. There are 8-9 estimated parameters are 10% significant, and 1% significant by likelihood ratio testing on total operation units. It means that there is significant difference among operation units. Therefore, it is inappropriate estimation by random profit function and should apply profit meta-frontier model for profit efficiency analysis.

We used LP and QP to estimate parameters in the profit meta-frontier model. Since both results are similarity, hereafter, we will only show empirical results  by LP estimation.

Because of operation units purchased paddy rice mostly/partially from contract production under the RPMA program, we can divide them into RPMA and non-RPMA with different production and marketing mode. It is assumed both modes have different profit performance and group profit efficiency frontier. Then, there is a frontier that is an envelope in both groups’ profit efficiency frontiers. The Group Profit Efficiency (GPE) is ratio of operation unit’s actual profit to optimal profit on their group profit frontier. Table 5 shows that the GPE of operation unit in RPMA and non-RPMA is 0.9100 and 0.7242, respectively. In other words, operation unit used non-RPMA mode with lower profit efficiency. It should do something in both increasing price and lowering cost to shorten the distance to optimal profit.

Table 5. Total and group profit efficiency estimation and testing

Note:*, **, *** represent 10%、5%、1% significance.

Source: the author’s estimation.

Since belonging to different group frontier, both group’s GPE cannot directly make comparison. We consider Meta Profit Efficiency (MPE) which is able to incorporate all operation units together. MPE is ratio of operation unit’s actual profit to potential profit on profit meta-frontier. The MPE is 0.6913 which means there is 30.87% of profit inefficiency to be improved. On the other hand, the MPE of operation unit in RPMA and non-RPMA is 0.8065 and 0.5760, respectively. Based on non-parametric Mann-Whitney testing, MPE in RPMA is significantly higher than in non-RPMA. Although contract price and processing cost are higher in RPMA than in non-RPMA, sale price is much higher in RPMA than in non-RPMA. The key factor is how to increase rice value-added. It could be a core strategy for developing the RPMA program in the future.

The MPE/GPE ratio also estimated for operation unit in RPMA and non-RPMA is 0.8863 and 0.7954, respectively. It indicates that Farmers’ Association and grain merchant both participating in the RPMA program could significantly improve their profit performance.

Determinants of profit efficiency

For exploring determinants of profit efficiency, it is supposed to be relevant to number of rice species, brands, channels, and channel types. We specified Tobit regression equation as follow:

Where y_it is a vector of GPE, MPE and GPE/MPE of i operation unit in t year, x_it is vector of rice species, brands, channels, and channel types, respectively. α_k are coefficients of relevant variables.   is error disturbance.

Regression results are shown in Table 6. We learned that variety of rice species, brand concentration and home delivery would enhance GPE of operation unit in the RPMA and non-RPMA program. In addition, variety of rice species and less brands of operation unit would improve its MPE and MPE/GPE. Those channels located in department stores, organic stores, and home delivery has better performance in terms of  profit efficiency. We can conclude that the RPMA program not only emphasize business model innovation, but also has to consider rice species, focus branding and choice of channel type in order to achieve the core objective, profit maximization, of operation unit.

Table 6.  Results of Tobin regression: determinants of GPE and MPE

Note: Number in parenthesis is standard error. Dummy variables for e-commerce, home delivery, department stores & organic stores, respectively. *, **, *** represent 10%、5%、1% significance.

Source: the author’s estimation.

CONCLUSIONS AND OUTLOOK

Rice industry is the most important industry in Taiwan’s agricultural sector. The government has been spending a lot by purchasing rice at guaranteed price, which causes financial burden, market mechanism distortion and resources misallocation. Under requirements of AMS reduction and competitiveness improvement, the government began implementing the Rice Production and Marketing Areas (RPMA) program in 2005. It rewarded operation units to make contact with rice farmers at a higher price than the government’s guaranteed price, and to run a business model by integrated value chain. The purposes of this study are to evaluate policy performance in terms of government expenditure and farmers’ income, and estimate profit efficiency of operation unit in the RPMA program.

The RPMA program has been in operations since 2005. It started with 27 operation units which included Farmers’ Associations and grain merchants and 4,908 hectares total, then increasing to 63 operation units with a total of 23,292 hectares total. Basically, the reward payment provided by the government is a key factor to attract Farmers’ Associations and grain merchants to participate in the RPMA program. The correlation coefficient is 0.86 between reward payment and participated hectares. Government could have a balance consideration on reward payment and purchasing expenditure saving to decide whether to increase the reward payment for participated hectares expansion or not.

Regarding the policy evaluation on the RPMA program, it has achieved the objectives of government’s purchasing paddy rice reduction. For example, we estimated reduction of government purchasing paddy rice at 46,584 MT and expenditure saving 1.1 billion in 2017. Government benefitted from a net expenditure saving of about 1 billion compared to 0.12 billion from reward payment. Besides, farmers’ income would be extra increased from contract price, which is higher than government guaranteed price by at least 0.1billin total. For instance, farmers’ income increased by 0.145 billion NTD total in 2017; in other words, each farmer’s income increased by 11,178 NTD average, or each hectare increased by 6,227 NTD average.

Participants of the RPMA program are government, farmers, operation units (i.e. Farmers’ Association and grain merchants). As analysis above, the program is beneficial to government’s expenditure saving and farmers’ income raising. However, the key factor is operation unit participation. If operation units would improve their efficiency and profit, it is expected that more operation units would participate in the RPMA program in the future.

We applied the Metafrontier model to estimate profit efficiency of operation units. Empirical results showed that the GPE of operation unit in RPMA and non-RPMA is 0.9100 and 0.7242, respectively. In other words, operation unit used non-RPMA mode with lower profit efficiency. It should do something in both increasing price and lowering cost to shorten the distance to optimal profit.

In addition, we consider Meta Profit Efficiency (MPE) which is able to incorporate all operation units together. MPE is ratio of operation unit’s actual profit to potential profit on profit meta-frontier. The MPE is 0.6913 which means there is 30.87% of profit inefficiency to be improved. On the other hand, the MPE of operation unit in RPMA and non-RPMA is 0.8065 and 0.5760, respectively. Based on non-parametric Mann-Whitney testing, MPE in RPMA is significantly higher than in non-RPMA. Although contract price and processing cost are higher in RPMA than in non-RPMA, sale price is much higher in RPMA than in non-RPMA. The key factor is how to increase rice value-added. It could be a core strategy for developing the RPMA program in the future.

This paper made regression on profit inefficiency, it showed that variety of rice species, brand concentration and home delivery would enhance GPE of operation unit in the RPMA and non-RPMA program. Those channels located in department stores, organic stores, and home delivery have better performance in terms of profit efficiency. We can conclude that the RPMA program not only emphasize business model innovation, but also has to consider rice species, focus branding and choice of channel type in order to achieve the core objective, profit maximization, of operation unit.

The RPMA program has been running for 14 years in Taiwan. Participation of grain merchants, farmers, and areas has constantly grown. However, participated area is only 8.5% of rice plant area. Since the program has effects of government purchasing paddy rice reduction, government expenditure saving and farmers’ income raise, government should actively encourage more grain merchants to participate in the program. Grain merchants will also improve its profit efficiency, build its own brand and have a better business performance after participating the program. We can say the RPMA program is a win-win-win program which is beneficial to government, farmers and grain merchants. Furthermore, products under the RPMA should be required by Taiwan Good Aquaculture Practice (TGPA), which ensures rice quality and safety. Actually, the RPMA program is also good for consumers.

From this year, the Taiwan government started implementing rice direct payment program, which is an alternative to government purchasing paddy at guaranteed price for farmers. This is if more rice farmers decide to take direct payment instead of selling to government at guaranteed price. The influence of government purchasing paddy at guaranteed price on purchased quantity and expenditure will decrease. We can expect that rice industry would move towards high quality and value-added development.

REFERENCES

Agriculture and Food Agency, Council of Agriculture (2017), Taiwan Food Statistics Book.

Battese, G. E., Rao, D. S. P. and O’Donnell, C. J.(2004), “A Metafrontier Production Function for Estimation of Technical Efficiencies and Technology Gaps for Firms Operating Under Different Technologies,” Journal of Productivity Analysis, 21(1), 91-103.

Chen, J. L. and M.H. Yang (2007), Study on Operation and Performance of Rice Production and Marketing Area, Project Report supported by the Agriculture and Food Agency, Council of Agriculture.

Eaton, E. and Shephered, A. W. (2001), “Contract Farming,” FAO Agricultural Services Bulletin, Rome.

O’Donnell, C. J., Battese, G. E. and Rao, D. S. P. (2008), “Metafrontier Frameworks for The Study of Firm-level Efficiencies and Technology Ratios,” Empirical Economics, 34(2), 231-255.

APPENDIX

Table A1. Estimation results of parameters and its standard error

Note: number in parenthesis is standard error, and *, **, *** represent 10%、5%、1% significance. Standard errors in LP and QP are estimated by Booth-strapping method.

Source: the author’s estimation.