THE CONTRIBUTION OF TURMERIC RESEARCH AND DEVELOPMENT IN THE ECONOMY OF BANGLADESH: AN EX-POST ANALYSIS

The study estimated the benefit and rates of returns to investment on turmeric research and development in Bangladesh. The Economic Surplus Model with ex-post analysis was used to determine the returns to investment and their distribution between the production and consumption. Several discounting techniques were also used to assess the efficiency of turmeric research. The adoption rate was found increasing trend over the period. The yield of BARI developed modern varieties of turmeric was 41 to 73% higher than those of the local variety. Society got net benefit Tk. 9333.88 million from the investment of turmeric research and extension. The net present value (NPV) and present value of research cost (PVRC) were estimated at Tk. 1200.84 and 157.88, respectively. The internal rate of return (IRR) and benefit cost ratio (BCR) were estimated to be 68% and 10.45, respectively indicated investment on turmeric research and development was a good and profitable investment. Seed production programme of turmeric should be taken largely to increase production by increasing area adoption.


Introduction
Turmeric is a spice derived from the rhizome of Curcuma longa L. which is used as condiment, flavouring and colouring agent, drug and cosmetic in addition to its use in social and religious ceremonies.It is a principal ingredient in the Bangladeshi kitchen as curry powder and paste.It adds flavour and colour to curries and has medicinal values also.In 1984-85, just before development and introducing of improved varieties of turmeric in Bangladesh yield per hectare was very low and it was only 1.92 tonnes per hectare.Yield per hectare of turmeric started to increase since 1985-86 and it stands 5.16 tonnes per hectare in 2007-08 production period, due to develop and dissemination of improved varieties of turmeric (BBS, 2012).
Due to increasing population, demand for cereal food increased significantly.To mitigate this demand, the land of turmeric crops is being diverted to cereal food crop cultivation.On the other hand, now-a-days many spice-processing industries such as Square, BD Foods, Pran, Archu, Advanced Chemical Industries (ACI), Amrita, Dekko etc. have been established in Bangladesh.They are exporting turmeric as a finished product outside the country.Due to that, demands for turmeric as raw material of these industries are increasing with the extension of their production.That is why; the total demands for turmeric are increasing at incremental rate.Resulting that, a big gap was observed between production and demand now.To meet up this gap the country has to spend a huge amount of foreign currency in every year for importing spices from abroad.
Realizing the importance of turmeric, Bangladesh government started turmeric research through Horticulture Division of BARI since 1980-81 for increasing turmeric production.But the fund was not sufficient to do research vastly.To keep the continuity of research of turmeric and other spices, Bangladesh government established Spices Research Centre (SRC) in 1994 under Bangladesh Agricultural Research Institute (BARI) for increasing the production of turmeric throughout the country.BARI has already released three improved turmeric varieties namely BARI Turmeric-1, 2, and 3.These varieties are cultivated in the farmers' fields since the release of these varieties.BARI, BARC (Bangladesh Agricultural Research Council) and DAE (Department of Agricultural Extension), to some extent have strengthened their works to turmeric production.However, for the research work of turmeric and its development, the contribution of BARC and DAE are greatly associated with BARI.
The present analysis thus took into the benefits from past turmeric research and its farm level development in the country.However, this study provided information for the policy makers, donors, researchers, extension people and the public on the contribution and the rate of return to investment in turmeric research in Bangladesh.

Sources of data
For the present study, data were collected from different sources like published and unpublished reports, and informal scientist's interview.The area, production and yield of BDMVs (BARI Developed Modern Varieties) of turmeric were collected from SRC (Spices Research Centre); adoption rates were collected through informal scientist's interview; and harvest price and consumer price index (CPI) were collected from various issues of Statistical Yearbooks (1985 to 2012) published by the Bangladesh Bureau of Statistics.The supply elasticity was taken from the study conducted by Dey and Norton (1993).Since SRC of BARI is the main organization for turmeric research, the research cost included mainly from SRC of BARI.The extension and promotion activities were done by DAE and the related costs were collected from this organization.BARC mainly provided the administrative costs.The on-farm yield data of BDMVs turmeric varieties were collected from the SRC, Bogra.Data on the input cost change was calculated by the researcher through analyzing increased production claimed higher labour costs for harvesting and transporting, expensive of seeds, and used slightly more fertilizers per hectare for improved variety than for traditional varieties.

Analytical procedure
The collected data were analyzed using the following statistical techniques.

Estimation of returns to investment
The Economic Surplus Model (ESM) with Ex-Post analysis was used to estimate the rate of returns to investment in turmeric research and extension.The analysis was done under small open-economy market situation.The theoretical concept of ESM has been illustrated below.

Theoretical concept of Economic Surplus Model (ESM):
The concept of economic surplus was used to measure economic welfare and the changes in economic welfare from policy and other interventions (Alston et al., 1995;Currie et al., 1971).Usually the economic surplus concept is adopted to estimate the benefits from the adoption of improved varieties.The components of economic surplus are consumer surplus and producer surplus.Given the initial condition (i.e., pre-research supply curve S1 and demand curve D1), consumer surplus is depicted as Area PoPnb in Fig. 1.This is the surplus or benefit to consumers because of a functioning market.Consumer surplus is that area beneath the demand curve less the cost of consumption.The cost of consumption is the area below the price line Pn.Producer surplus is defined by area PnbO in Fig. 1.Area PnbO in the surplus left to the farmers after they have paid for the total costs of production, area ObQn (Alston et al., 1995).
The adoption of an intervention by farmers, such as an improved variety usually means one of two things: (i) A farmer can supply more of the commodity using the same level of resources (i.e, same land area and other inputs), or (ii) A farmer can supply the same level of commodity output but do it with fewer resources.
In either case, this is depicted by a shift to the right of the supply curve as shown in Fig. 1 (the shift is from S1 to S2).The shift is the supply curve from the adoption of an intervention changes the initial equilibrium price and quantity of the commodity.This new price quantity equilibrium increases economic surplus.The change in economic surplus (economic benefits) is measured by comparing the difference in economic surplus between the pre-adoption period and the post-adoption period.

Change in Consumer Surplus
= Area abc + Area PnbaPo Change in Producer Surplus = Area Oac-Area PnbaPo Change in Total Economic Surplus = Area abc+ Area Oac Given a shift in the supply curve S1 to S2, the change in consumer surplus is depicted in Fig. 1 as Area abc + Area PnbaP0.The shift in the supply curve (due to the adoption of an intervention) has decreased the price consumers now have to pay for the commodity.
Given a shift in the supply curve S1 to S2, the change in producer surplus is depicted in Fig. 1 as Area Oac-Area PnbaP0.Area Oac represents the decrease in the cost of production the same unit of the commodity that farmers now enjoy because they are using the intervention.This represents the benefits to the farmers from adopting the intervention and can be measured and quantified in monetary terms.The adoption of the intervention, however, has increased the quantity produced thereby decreasing the price of the commodity (Pn to P0 in Fig. 1) and is a loss to farmers income.Farmers can recover some of this loss since they can sell more quantity (Qn to Q0 in Fig. 1) of the commodity.
The total social benefits to society from the adoption of an intervention is the summation of the change in consumer surplus plus the change in producer surplus (Area abc + Area Oac) minus the input cost change from adopting the new interventions.
For a closed economy model, the estimated price elasticity of demand is used in the above formulas.For small open-economy model where the elasticity of demand is perfectly elastic, use a sufficiently large number of η (Nagy and Alam,   2000).A small open economy market is one where the amount of exports or imports is small relative to total world trade in the commodity.Thus, there is little or no effect on the world price of the commodity (the small country assumption).
In this case, the price of commodity does not change with the shift in the supply curve.For this study, the Bangladesh turmeric market is modelled as a small open economy market.
Fig. 2. Small Open-economy Importer Economic Surplus Model The change in economic surplus for a small openeconomy that is domestically produced but allows imports to cover shortfall (i.e., the Bangladesh turmeric market) is depicted Fig. 2. The world price Pw and quantity demanded by Bangladeshi consumers Q1 defines the initial equilibrium.At price Pw, producers supply Qn amount of turmeric when faced by the pre-research supply curve S1.Turmeric imports are equal to QTn.When faced by the research induced supply curve S2 (the supply curve that exist because farmers have adopted new high yielding varieties).
Turmeric producers increased production to quantity Qn and increase QnQ0.Spices imports are decreased by the same amount as the increase in production QnQ0 and are now at QT0.Since Pw does not change (small economy assumption), there is no change in consumer surplus-consumers are neither better off nor worse off.The enter change in economic surplus from the adoption of new turmeric varieties is thus a change in producer surplus only and is identified by area oab in Fig. 2 (corresponds to area oac in Fig. 1).The amount of foreign exchange saved by the adoption of improved varieties is equal to Pw x (QnQ0).
Empirical approach: The Akino and Hayami (1975) approximation formulas for calculating changes to producer and consumer economic surplus are described below and these are used in this study.The Akino and Hayami (1975) approximation formulas for calculating the change in economic surplus for a closed economy analysis (Fig. 1) is as follows: The supply shifter (k): The supply shifter 'k' is the overall yield advantage of improved varieties of turmeric over the local variety weighed by the area sown to the improved varieties of turmeric.
In the case of the Akino and Hayami (1975) approximation formulas, k is the horizontal shift from the equilibrium price Pn given S1 to the equilibrium price Po given S2 which corresponds to a distance equal to QnQo in Fig. 1 (Gardiner et al., 1986;Nagy and Furtan, 1978).
The supply shifter k is calculated as follows: Where, Yit = Yield of the improve varieties of turmeric in year t Yt = The yield of a base (or average yield of local variety turmeric) that has been gown in the past and that would still be grown if no new varieties had been developed Ait = The proportion of the total area sown to improved varieties of turmeric in year t n = The number of improved turmeric varieties

Estimation of net present value (NPV)
The amount of total funds returned from the investment in research is called NPV.The benefits were calculated by using the following formula of NPV: The cost of research and extension investment in year t r = The discount rate n = The time horizon over which the benefits of the research investments are realized

Internal rate of return (IRR)
The IRR was calculated relating to the total social benefit (TSB) minus an input cost change, if any, in each year to the research expenditure (C) in each year and is the discount rate that results in a zero net present value of the benefits.The IRR is calculated as The IRR can be defined as the rate of interest that makes the accumulated present value of the flow of costs equal to the discounted present value of the flow of returns, at a given point in time (Peterson, 1971).

Adoption status and yield advantages of BARI developed modern varieties of turmeric
The adoption of improved variety is very important factors by which the volume of change in economic surplus is determined.The more the adoption of improved varieties over traditional one, higher the change in surplus will be.Apart from this, it gives us feedback as to why and how well a technology is being accepted by the farmers.There was no turmeric varietal adoption survey conducted in Bangladesh.The existing variety survey information along with the considerable field experience of the spices scientists is used to sketch out the per centage area sown by variety grouping which are presented in Table 1.

Supply shifter k
The supply shifter k identifies the amount of production that can be attributed to the varietals improvement research in each year (i.e., the shift in the supply curve).The more the value of supply shifter the more is the shift in the supply curve, resulting higher benefit to the society.The supply shifter is the outcome of the simultaneous force of adoption per centage and yield advantage.It was calculated using the formula discussed in methodology.Yield advantages: This is very important factor to determine the economic surplus.The higher yield advantage always ensures higher level of economic surplus.Two types of data exist in most of the less developed countries for good estimation of yield advantage (YA) as well as the aggregate production function shifter.They are on-station yield trial data and on-farm yield data.The on-station yield data is readily available and most often the only reliable source.One of the arguments against using on-station yield trial data is that of superior management practices and techniques are used and therefore, the results may not reflect on the on-farm situation.Another argument placed by different author (Hertford and Schmits, 1971;Ayer and Schuh, 1972;Akino and Hayami, 1975;Scobie and Posada, 1977;Nagy and Alam, 2000) and showed that the yield advantage estimation from the on-station yield trial data would be biased upward because the estimation might also include the contribution made by inputs such as fertilizer and water.To account for this problem, the estimated yield advantage of new varieties by estimating production functions of yield as a function of new varieties and other inputs.This process requires a substantial data which is not readily available in Bangladesh.
For the present study, on-farm yield trial data were considered as a more reliable source for the calculation of yield advantage rather than the onstation yield data in Bangladesh.The yield advantages have been calculated for this study following Gardiner et al. (1986), Nagy andFurtan (1978) andNagy (1991).
Developed varieties of Spices Research Centre (SRC) of BARI have replaced the traditional varieties starting in 1985-86 for BARI Turmeric-1 and BARI Turmeric-2 and 1999-00 for BARI Turmeric-3.The weighted yields were calculated by taking the average of the irrigated optimum, late irrigated and non irrigated yield multiplied by the mean of irrigated, late irrigated and non irrigated area of turmeric.Per hectare average yield of high yielding varieties of turmeric i.e.BARI Turmeric-1, 2 and 3 were found to be 4.12, 2.95 and 6.48 tonnes, respectively.In case of traditional variety it was only 1.74 tonnes.Therefore, the yield advantages of BARI Turmeric-1, 2 and 3 over traditional variety were found to be 58, 41 and 73%, respectively (Table 4).

Estimating benefits from turmeric research and extension
This section deals with the estimation of returns to investment in turmeric research and extension using the economic surplus approach.This approach estimates the benefits to agricultural research by measuring the change in consumers' surplus (CS) and producers' surplus (PS) from a rightward shift in the supply curve that is brought about through technological change.It should be mentioned here that aggregate consumers' surplus, producers' surplus and total surplus were calculated by summing up corresponding surpluses of all turmeric rather than summing up from the areas of the model.In order to calculate the net benefits (NB) research and extension expenditures are subtracted from total surplus.All these estimates of benefits are expressed in real term by using 2007-08 constant prices.The rates of return and NB are then discounted using 10% interest rate for obtaining the efficiency of investment.First, the yearly total social benefits are estimated using the small-open economy model (Fig. 2).This is done by assigning a very high number to the demand elasticity parameter (ŋ) since in a small open-economy model, ŋ is perfectly elastic.The analysis is undertaken for each year 1980-81 to 2007-08 for turmeric.
Turmeric research and extension in Bangladesh are seemed to be continued by three different organizations.The Organizations are Bangladesh Agricultural Research Institute (BARI), Bangladesh Agricultural Research Council (BARC) and Department of Agricultural Extension (DAE).The turmeric research and extension expenditure comprised the expenditure of three organizations are furnished in the following sequence.
The year wise expenditures behind variety development and dissemination for the new varieties to the farmers of turmeric are shown in Table 5.The expenditures of BARI/SRC and BARC were estimated from 1980-81 to 2007-08.The accumulated expenditures over the years of BARI/SRC and BARC were estimated at Tk. 95.88 and 17.71 million, respectively.Extension expenditures and input cost change were estimated after development of improved variety and they were started since 1985-86.The cumulative expenditures of extension and input cost changes were respectively amounted at Tk. 67.02 and 530.04 million.Over the years, expenditures accruing for BARI/SRC, BARC, DAE and input cost change were Tk.710.65 million.For the analysis, the current total expenditures were converted to 2007-08 constant prices using the Bangladesh Middle Income Group CPI Index and it was Tk. 987.70.6).

Rate of return to turmeric research and extension
The rates of returns are the indicators which help to estimate the investment efficiency of the research programme.There are many types of measures that can be used to estimate the rates of return.The benefit cost ratios were found to be 10.45 for turmeric.The value of the parameter clearly indicated that the investment in research and extension of turmeric in Bangladesh is a good investment and highly profitable.

Foreign exchange savings
The yearly increase in production due to research save the country's foreign exchange to a remarkable extends.First, the research induced productions for turmeric for the past years were calculated by multiplying the country's total turmeric production by their respective production function shifter k.Multiplying the results by world turmeric price, foreign exchange savings was obtained.

Policy implications
The empirical results indicate that the expenditure on turmeric research and development paid a favourable rate of returns and the society were also benefited enormously out of it.The IRR to turmeric research and development expenditure was found to be 68%.A 68% IRR on investment in turmeric research and development is a good rate of return.The consumer's surplus is found to be very few only due to small open economy.But this situation might not be the good sign for the economic prosperity.For the survival of the consumer, price support should be given by government.
The annual adoption rates of BARI Turmeric-1, 2 and 3 are not good because of non availability of seed.Seed production programme should be taken largely by the government and nongovernmental organization so that the farmers can get quality seed easily with a reasonable price.

Price
of turmeric (Tk./ton) (Existing market price) Qo = Production of BDMVs turmeric (ton) (Existing production) Pn = Quantity price that would exist in absence of research Qn = Quantity of the turmeric produced that would exist in absence of research k = Horizontal supply shifter γ = Price elasticity of turmeric supply η = Absolute price elasticity of the demand for the commodity.

Fig. 3 .
Fig. 3. Foreign exchange savings due to turmeric research and development over time

Table 1 .
Area of traditional variety replaced by BMVs of turmeric Source: Several issues of BBS Note: BDMVs-BARI Developed Modern Varieties, LVs-Traditional Varieties and shaded area indicates no improved varieties were released.

Table 2 .
Adoption rate of BARI developed modern varieties of turmeric

Table 3 .
Calculation of the supply shifter (K) of BARI Turmeric over traditional variety

Table 4 .
Yield advantages of improved varieties of turmeric over traditional varieties

Table 5 .
Turmeric research and extension expenditures by sources from 1980-81 to 2007-08

Table 6 .
Estimation of surplus from turmeric research and extension investments

Table 7 .
Estimated rates of returns to turmeric research and extension

Table 8 .
Foreign exchange savings from investment in turmeric research