Policy, Research, and External Affairs WORKING PAPERS International Trade International Economics Department The World Bank M3rch 1991 WPS 607 Abolishing Green Rates The Effects on Cereals, Sugar, and Oilseeds in West Germany Conald F. Larson Simon Glance Brent Borrell Merlinda lngco and Jonathan Coleman Eliminating the price difterenitials that result from country- and corrmnmodity-specific exchange rates ("green rates") would re- duce farm inicomie and devalLe fixed agriCultural assets. This complicates thie difticult task of reform that is essential if there is to be a unified European market. I ' I' w1R . * Ix, \ s .S v . > k . i'l A k N1.x '.7 I d< !*{ 1.o > prozgrcss an,! \I '.. V.: * ' 'awr~ 1'' SBc li ? ' U '' '''" '' ' ' 'i"' "' lCS (Ii d ISr T Q Qc rlarncs 0l . , . .' " , . ' ./'Trt' d .':\ .~r' t4 i!'.I ar.l o?.o. '. A t . . ' . r' ' .~~~~~~~~~1- ' ' L .'. I' 'II ' f !5' t.' (..... "'u.' r"i: "' r r o > nc-r Ie.Ul 11iilmunil. cliiinliiin" r' price ditfe'reces resulline \u2eNCSts tllt Cl1 imitltaine 2rCen atu1 FC u \%OLIl Claid 1I0ro1 COLotno-\ - and c:ollllllomdit\ -sspecilic e'Challn ptinimarl\ to a dec'lin in a,nnr inicoImec atnd at ra:teNi', k Ir I Jn a- ir 'cl C rles. dI luI[tion ofl fie\Cd arl-itLLil ItN I hicli I CU plkni:lica the dilldiffiull t4.ik of ellillilln " I'.'ill I .;il'(iallnd Iis a\sSOeiates (diSuNN te1C \ nliNfLl 1li1\ illNtliUil tieit 1tha t eoullicate t1it eifetN l O I.I .\\ITet'i ::;fifiku .\E .,l;\; \aI;'P\<s> 8itt!Ets,ilus.Set,il l..,,xi>,,.!liKi~~~~~~~~~~ Abolhbing GRate Rs: The Effects on Cereals, Sugar, and Oilheeds in Wes Germny by Donald F. Lason, Simon Glance, Bret Borel, MerUnda Igco, and Jonathan Codem Summary In 1987 the E4iroF : Community began the ambitious task of forging a single market for goods and services across the nation? srders of member states by 1992. Substantive reforms to the Community's Common Agricultural Policy, necessary for a full integration of existing markets remain to be accomplished and have proven difficult to aihie%e. The creation of a truly "common" agricultural policy in the EC requires, at a minimum, th, elimination of price differences resulting from country-and-commodity-specific exchange rates, known as "green rates". This paper discusses the variety of policy instruments which complicate the effects of these policy-determined prices differences on crop production and the demand for inputs. A model is presented and estimated which measures both the significant cross-commodity biases created by the multiple policy instruments and quantifies the effects of removing green-rate differentials in (formerly) West Germany. While the effects of prices changes on domestic production are statistically significant in the model, they are shown to be quantitatively small. Such a result suggests that eliminating green rates would lead primarily to a fall in farm income and a devaluation of agicultural fixed-assets, exacerbating the difficult task of attaining reform. Aboli:.hing Green Rates: The Effects cn Cereals, Sugar, and Oilseeds in West Gerlany by Donald F. Larson, Simon Glance, Brent Borell, Merlinda Ingco, and Jonathan Coleman Table of Contents Introduction 1 Intervention Instruments and EC 1992 1 Cereals and variable levies 2 Sugar quotas 2 cilseeds and the crushing subsidy 2 Profit, Supply, and Demand Effects of Common Aglicultural Prices 3 Empirical Model 6 Estimation Results 7 Jointness in Production 10 Dynamic Production 10 The Significance of Pricing Policies 11 Measures of Bias in Resource Allocation 12 The Effects of Abolishing Green Rates 12 Concluding Remarks 15 Appendix A: The Effects of a Currency Realignment on 16 Relative Prices Appendix B: Calculation of Monetary Differential Amounts 17 Bibliography 19 The .fecu of"aiAkg gr^. swa Lame. d,. a. Intoduction Tbe cieation of a borderless market in the Economic Community requires, at a minimum, that bomogeneous goods be priced the sune regardless of wbere they are pr('duced. For agricultural goods, this requires, among other things, the elimination of "gee rates" and derivaive policy interventions such as MCAs and MDAs. Because of the multiple ways in which the EC intervenes in agriculture, eliminating green rates would not have a uniform impact on all craoj. It is agued that if green rte wae elmnted, relative prices would change only across broad dasses of agrcutura goods, such - ceal and oilseeds, but nt wiin tdwee categod. Since conunmer subtitotion unongd the major I cops coveaed by this paper-cereals, odseeds, ua sugr- is imal ac them aggrepte groups, the pimary t _pon to elimnating green rates will come fom poducers. To the extent that pduicion is joint, government poL;cies wbich affect one crop directly create seconday effects in otber markets; eiminating greeo rates will rte coresponding da d indirect effects as well. In this paper, a general restricsedprofit modl is psed and estimated Resolts Sprt the notion ta duction iS joit and measurs of bia resulting from fixed facton ad fom sugar qot are quantifed. Estmaton eWslts indicate that prices ae saly imptat in the sbort-uo, but mulation results demonrate that the supply and input-demand effects of eliminating green rts are, in the sbort-run quanttively small in West Germany. Interw "o isuruments and C 1992 In June, 1985, the European Cowmssion issued a White Paper enitied "Completing the kItera Maket." The paer oudined 279 directe, wbich, if imp_emed would crate a European matket without borders. One hunded of the diectives dealt with agriculture. One month later, the Commission relased a Greeo Paper entitled "Perspectives for the Common Agricultral Policy," which provided a major review of the CAP and the escalating costs of the policy progmns. In 1987, the Single European Act which amended the Treaty of Rome to enable the adoption of the refonn package was nrafied by all membes; and in Febnuary, 1988 an agrement was reached at the Bmssels Summit to finance the reforms to compietion by the end of 1992. Article 8A of the Single European Ac states that: lbe Community haa adopt msm r ith the aim of progeesvdy establishing the inthnal mrki:' over a period expiring on 31 December 1992.Me intenal market shall comprise an area witout inenal lrntis in whihthe free movemnt of goods, persom, services and capil is e red In acoord withte provion ofthsTreaty. Kelch (1989) points out that an EC without borde has four fundnmental implicatioos for EC food and agriculture: 1) the hannonotion of plant ad animal health sandards, and food labeling, ingredients, and packaging laws; 2) the harmonization of taxes on food and agricuhlu prducts and inputs; 3) the elimination of agricultural bord taxes uand subsidies; 4) temoval of quota, variable premiums, and rational aids which are incompatible with the 1992 program. Impl t of these refonms would brng about a single price for agricultural goods for consumars and produces throughout the Community and constitute a rvision of relative and absolute pnces anong commodities and countries. It is agued below that. in general, the proposed cbhnges are likely to affect the decision- making processes of producers more drUcaly than those of consumers, leading to a greater alloction of rsources in prodaco than in demand. The result stems from the multiple instruments by which agricultural prices are suppoted under the Common Agricultural Policy. For the major aneudl crops, thre major progams exist. Cereal producers and consumers face a legislated price substantially above world levels which is defended by a variable levy system. Sugar producers receive quotas which enable them to sell a fixed mount of their production at legislated prices higher than world prices, with the I The effects f abolahjg greem rates Larsor. #t. al remainu.er to be exporte. At world pnces. Oilseed producers receive aid in the form of an indirect payment to oilsc:d crushers, due to a binding trade agreement entered during the Dillon-Round of GATT in 1962. Cereals and variable levies: The variable levy program is conceptually straight-forward but, in practice, it is complicated by exceptions and modifications. For example, in order to remove supplies which would otherwise dictate lower consumer and producer prices, cereals are exported with a subsidy roughly equal to the difference between world prices and domestic EC prices. As world prices change, the subsidy, rather than domestic prices, changes. In the case of crops for which the EC is not self-sufficient, importers face a variable import-tax equal to the difference between the intemal legislated EC price and world prices, preventing consumers from substituting relatively inexpensive imports for expensive domestic products. The result ot this intervention was to create a single, higher and more stable farm-gate price (and consumer price) within the borders of the EC. This simple mechanism was greatly complicated in 1969. Obtaining a common level of farm suppon was a stated objective of the CAP at the inception of the EC and much progress had been m'Ale prior to that date in removing border-taxes, but a devaluation of the French franc and the revaluation of the German mark in 1969 brought about a major revision in the way in which cereal support levels were calculated. Unwilling to let CAP rupport prices rise by the 12.5% implied by the devaluation, French authorities continued to use the exchange rate established prior to the devaluation when calculating commodity support levels. The Gerinans, unwilling to accept an 8.5% cut in support prices, used older exchange rates as well--and the practice of using "green" exchaDge rates (i.e., specifically to prevent common agricultural prices) was established. Since support prices in France were lower than those in other Community countries a~d cousiderably lower than those in Germany in particular, free trade across member borders would have driven up prices in Prance and lowered those in Germany--exactly what France and Germany intended to avoid. To prevent this, an amount equal to the difference (either positive or negative) vwas levied against trade crossing French or Gennan borders, thus neutralizing the legal gains to trade created by the use of artificial exchange rates. These levies becane known as monetary compensatory amounts or MCAs. Since that time, green rates have proliferated; there are currently 40 green rates among the 12 members of the EC (Kelch: 1989). Italy, for example, has one green rate for grains and oilseeds, ancther for sugar, peas, and beans, a third for pork, and a fourth for beef and milk. Sugar quotas: The EC support mechanism for sugar is based on a complicated system of quotas, nrles, legislated prices, and import levies. There are thlree categories of sugar: A, B, and C. ProJ1uctioa of both A and B sugar is limited by quota. Both receive a support price well above world market levels which has helped transform the EC from a net importer to a substantial net exporter of sugar. The price support for A sugar is above that of B sugar and A sugar constitutes the bulk of production. C sugar is considered excess production, cannot be sold in the EC, and must be exported at international prices. Producer support pnces are for A and B sugar ame set in ECU terms and then transformed to local currencies using green rates. The green rate used for most countries is the same rate that is used for cereals and oilseeds; however, as stated above, Italy uses a separate rate tor sugar, peas and beans. The green rate used for sugar support conversion in Spain i the same as that used for cereals, but differs from the oilsted green rate. A series of import taxes prevents sugar imports from all countries except those that entei under the Lome Agreement. OUseeds and the crushing subsidy: The support mechanism for oilseeds is perhaps the most convoluted. It has its antecedents in1 a concession made by the EC durin4, the 1962 Dillon round negotiations on the Geneal Agreement on Tariffs and Trade (GATT) to allow unrestricted imports of soybeans and to limit the tariff on vegetable oils and meals. At that time, the EC maintained a large oilseed crushing industry but oilseed production was negligible. During the 1970s, as the FC moved toward a policy of obtaining self-sufficiency in oilseeds, the 1962 concession prevented the establishment of a variable-levy system similar to that used in cereals. In response, the EC established a system of production aids, or crushing subsidies which are paid to crushers of domestic oilseeds. Rules guarantee that a large portion of the payment is remitted to farmers. In addition, the EC stands ready to buy the oilseeds at a minimun intervention price. The program ensures EC oilseed producers of a price substantially 2 The iffac '5 eof atlishg gr#en rao;*. LArso, #I. al. above world levels, and has been quite effective in expanding oilseed productioo.' The indirect natur of the program, however, has created an additional set rf listortions whereby relative prices of cereals and oilseeds can differ from country to country even when cereal a)d oilseed prices are both con. ned by a single green rate. This obscure distortion occurs because the support comes indirectly via a crushing subt dy. Oilseed crushers ae free to import and face world market prices. In order to encourage crushers to buy t. mnore expensive EC produced oilseeds, the EC calculates the difference between the world price in ECUs and the suppont price in ECUs. Sine the world price changes daily the subsidy fiu.tuates as well. The difference is converted using "green" exchange rates and paid to the crusher, who passes the bulk of the subsidy back to producers. The fanmer, in effect, receives the world price --which is converted at the official exchange rate-- plus an ECU-denominated bonus via the crusher - -which is converted at the green rate. While the cereal support price is fuUly converted using the green exchange rates, the farm-price for oilseeds is a strange combination of world prices, official exchange rates, subsidies, and green rates. Hence, the price of oilseeds relative to cereals will differ from country to counry. Annex A gives a numeric example. The use of multiple exchange rates introduces artificial incentives for trade in oilseeds as wel as cemeals. However, since the support price is a weighted sum of international prices converted by one set of exchange rates, and a production aid converted by another, the corresponding mechanism used to calculate taxes or subsidies at the border of each member nation is more complicated than the MCAs. The Monetary Differential Amounts (or MDAs) are calculated so that the cost of the oilseeds are identical to crusbeis located throughout the Fw despite differing levels of producer support. Since the market exchange rate may fluctuate as well as the market rate for oilseeds, the MDAs are also in a constant state of revision and adjustment. Annex B gives a numeric example. Profit, supply, and demand effects of common agricultural prices The creation of a borderless EC requires, at a minimum, the eliminaticn of separate green rates and their derivative instruments, MCAs and MDAs. Since the current system of multiple exchange rates and production aids creates country differences in both absolute price levels and relative ptices, moving to a single set of agricultural prices will have real effects on consumption and production. Consider first the effects on profits in agriculture, and therefore on resource allocation between agriculture and the rest of the economy. Let x = a row vector of exchange rate ratios on output prices, and y = a row vector of exchange rate ratios on input prices, where member A = e,'/e and e,r is the EC!T!country exchange rate used for good i, and e is the market exchange rate. Noting that some of the A, and most of the y, may equal one, the aggregate profit function can be written as: xQ(6p,yw,q,z) = Apy(-)-yw() (1) where q is a vector of the production quantities for supply-managed crops such as sugar and olive oil, where output prices (p) and input prices (w) are intervention prices stated in ECUs, and z is the vector of fixed inputs. Under a borderless EC, A=y,=l, for all outputs and inputs so that the relative, as well as absolute prices will change. For example, in W. Germany. where the green rates over-value support prices (see Figure 1), setting 'While the creation of a single-market Europe in 1992 promises potentially large changes in all support programs, the future of the oilseed regime in the EC remains even more clouded. In 1989 the EC accepted a GATT ruling in response to charges brought by the United States that the crushing subsidies violated the 1962 agreement and had hampered soybean exports from the US into the EC. In response to the ruling, the EC must either alter the support program. pay damnages to the US. or negotiate an alternative setlement. Th~ q <.btJ z L. a.gf F Ratio of green/market exchange rates for selected EC nember countries 12 a 0 6 _ VII' 0. 4 1979 1961 1983 1966 19&2 196 196 1966 Figure 1: Ratio of gmen-to-market exchange raes for cesa1s in selected C couuies. the ratio of green-to-market exdhange rates to I would significalty lower all producer output psices. Generally speaking, green rates do not affect input prices direcdy. A notable exception is in the lvestock sector wvrre production from the annual crop sector is used as inputs to both dairy and meat production. A notable indirect effect is the influence that support prograns have on land values.2 To the extent that the inputs are tradeable, a borderlss EC may lead to a reahigmemt of tax rates on inputs as well, as farmers will seek the lowest cost inputs. Augmeting y to inchide diffeences in effective tax rues, the effects on supply and input demand czn be derived using the regular envelope theowm properies, so that: t/cUp - $Lp,yw,q,g) (2) Pnd - -x(%p,yw,q,i), (3) where y( ) is a vector of shon-run supply curves, and x(') is a vector of short-nm input demand curves. From (2) and (3), it is clear that a borderless EC with a single real price structure, wbere 1-,-l will lead Mnhe model developed in this paper is limited to the major annual crops only and cannot address these issues. However, the effects of prices changes on the livestock sector and land prices are discussed in an fonhcoming extension of the model by Ingco and Larson. 4 TA. .jecs of aboljhing green rag.s Larson, et al. to a change in both supply andi input demand even in the short run when fixed resou: ;es cannot be reallocated. To the extent that most of the r, are equal to one, supplies should decline (increase) in countries where the ratio of green-to-market exchange rates are greater (less) than one, since the supply curves are monotonically increasing in p. Input demand should decline (increase) as well. Equation 4 provides the shadow price on the fixed inputs and provides the direction of the long-run adjustment: F(-kP'YW'v1 ~~~~~~~~~~(4) a3 i r(A.p,yw,q, 8'4 To the extent that the value of the shadow pnce is reduced below (increased above) the marginal cost of the fixed input when ,yl, the utilization of the inputs which are fixed in the short-run, wiU be reduced (increased) in the long-run as these inputs become vaiable. The cbange in absolute and relative prices will affect final consumer demand as well; however, the magnitude of the changes is expected to be more limited than the supply effects due to the more limited opportunities for substitution. Consider the utility maximization problem: Max U = U(s a.t. 7 Apd - B(S where d is a vector of consumption levels and B is the budget constraint. It is clear from (5) that a change in A wiU bring about a cbange in consumption levels thrugh the budget constraint, and, because of relative price changes, lead to a price effect on demand. However, to the extent that groups of commodities are separable--that is, to the exte. k that consumption of cereals is independent of the price of sugar, or vegetable oils and meals--these effects should be quite limited. For the purposes of this paper, total budget expenditures is divided into expenditures on sugar, vegetable oils, cereals, and other goods. While the demand-elasticities of substitution among vegetable oils are quite high, under current policies consumers face intemational price levels that are independent of support prices. Changes in taxes are likely to affect all vegetable oils eqLally, so relative prices among vegetable oils are not likely to change. The substitution among cereals may be significant and support prices do effectively detenrnine consumer prices. However, since all cereals face the same green rates within each country, green rate changes will again not affect relative prices. Sugar is a single product and sometime- faces a green rate different from that applied to cereals. To the extent that relative prices within the three groups do not change, the only price effects of a price change on final demand will come through the substitution among the groups and the income effects. The demand cross-elasticities for sugar, vegetable oils, and cereals are likely to be small if not zero. In wealthy nations where a small porti,n of income is spent on cereals, vegetable e., . and sugar, the income effects are likely to be small as well. Qualitatively, the move to a unified market and the abolition of anificial country-specific prices for homogeneous goods should have the following effects: I) small changes in final demand with little substitution effects. 2) changes in resource allocations devoted to agriculture among countries, as well as a reallocation of resources within countries among crops. 3) changes in the value of non-t aded inputs, especiaUly land. The remainder of the paper deals with quantifying result 2 for the major annual crops, cereals, oilseeds, and sugar in (formerly) West Germany. lbe effects of the proposed policy changes have potentially large and interesting 5 ThA effi 15 f ab&lishng g,em jIfT Its 'A rt effects on non-traded assets such zLs land and ol(der tarmers, howAver, ihe etfects on resource adiocatioki and the resulting supplies ot' these goods will potentially have the greatest effects oi international markets and other, panicularly developiig, countines Empirical Niodel Appl:catioiLs based on duality result. and flexible-torin estimates of jointly produced agricultural products have flourished in rpcent years and include Ray (1982). Shumway (1983), .. .pez (1984) atnd Lee and Chambers (1986). The implications of supply-managed outputs, in addition to fixed inputs, was first rnsidered by Moschini (1988) who showed that the restricted profit function given in (1) is nondecreasing in p and in z, nonincreasing in w and in q, positively linearly homogenous in (p,w), convex in (p,w), continu is, and twice-differentiable. Because of the constrained output vector q, profits need not be positive. Using the SPEL database, provided by the University of Bonn, flexible output and input-demana groups were created for cereals (barley, oats, maize, rye, and wheat), oilseeds (rapeseed), non-quota C sugar, ferti!izers (nitrogenous, phosphatic, and potassic), energy, and pesticiles; the following groups were designated as fixed outputs or inputs: quota-sugar (A-quota plus B-quota), ne' subsio.s ksubsidies minus taxes), depreciation (on machinery and buildings), other inputs (based on total input constar -price costs minus explicitly modeled expenditures, and comprised primarily of maintanence expenditures). The data covered the period 1967 to 1987. Aggregate quantity and price measures were created using divisia indices. Following Moschini (1988), the normalized quadratic, first proposed by Lau (1974) and subsequently applied in profit function estimation by Shumway (1983), was chosen as the postulated f.inctional form. The form allows for negative profits, and has a Hessian of constants so that global convexity can be imposed and tested. Choosing one of the inpuits as a numeraire (we chose energy), let 4 * (y-~x1,...... -X) ... (6) represent the netput vector corresponding to the normalized prices *n (p,w,,...w,_), (7) obtained by deflating each price by w,. Defining f* (q,z,t), (8) where t is a trend variable representing the state of technology, the normalized quadratic restricted profit function is written as: r + + 2 (9) where i,m = 1,..., M and j,n = 1_...,N; whet.. x, c*, o, 3p. re fixed parameters. Note that profits are normalized, that is: * Xfi st (p,w,qgz) ( 10) wJ so that rl(p,w,q,z) is linearly homogenous. Symmetry is imposed by setting o.n _ cX,n, and ( = h 7it, ef a ,,hshin iii,,&n 'iat Larxim to, al, The nlcput tuinictions can be recovered via the envelope theorem: aft~ ci b.P (I11) for each of ihte 11 variable netputs, Ideally the M flexible netput equations given in ( 1) are estimated together with the profit function given in (10). However, including the profit function adds an additional set of parameters corresponding to 1 and p3 leaving the empincal problem unmanageable. Because of this dilficulty, oaly the M netput equations were estimated This procedure still allows price and cross-price elasticities to be estimated for all variable outputs and inputs as well as parameters measuring the marginal effects on constrained outputs and inputs on variable netputs. Lau (1978b) first proposed a nonlinear transformation of the equations in ( 11), also used by Moschini, which allows the property of convexity to be directly tested. Letting A represent the M x M matrix of the ct, coefficients, the restricted profit function will be convex if A is positive semi-definite. Since A is symmetric and square, it can be decomposed so that A = LDL', where L is a unit lower triangular matrix and D is a diagonal matrix. The elements of D are called the Cholesky values which must be nonnegative for A to be positive semi-definite. Cholesky factorization renders the equations in 11) nonlinear in their parameters and stochastic versions cf the netput e-quations as well zs their correspor.iing "Cholesky" versions were estimated using a Maximum Likelihood procedure Justification for this technique is given in Amemiya (1983), Chalfant and Gallant (1925), Moschini (1988), and Weaver (f83), Estimation Results The first two' columns of Taole I provide the summary statistics for the model estimated in its base-form. The underly0ie profit-tunction is linearly homogenous in prices by construwtion and symmetry has been imposed. Pro(Jiciti is assumed to be jcillt. Convexity of the underlying profit function has not been imposed at this point. T* suniniarv statistics are fair;v go(xi, but not greatly revealing. The explanatory power of the equations are umilorriil gow(A. and only the equation for oilseeds appears to be -.fccted by first-order autocorrelation. 'T'able I: Summary statistics for estimated normaliz&d quadiat-c restriced profit function with and without convexity constraint. Deperndent Adjusted Durbin- Adjusted Durbin- Vanable R- Watson R '0Watson Unconstrained Nlodel Constrained Model (Grains 0.84 2.22 0.82 1.94 Oilseeds (099 3.23 0.99 3.04 C-Sugar 0.83 2.09 0.84 2.13 Feritihzer 0.95 2.00 0.95 1.74 Pesticides o098 2.20 0.97 1.68 Nle.iu-point price and quantity elasticities and the associated asyniptotic t-scores calculated from the model The offcu LI yabolshrig gremn rates Larro0. et. al. Table 2 Elasticity estimates at the mean point from estimated normalized quadratic restricted profit model. -- with respect to the M1cC of wid- mepect to the quadty of---- Quota Otor Flesiorv It Cereals llseeds ('..ugSr Ferftltzer Peuticides Eaergy Susr Capital Time Inpuab S.ubtdis LAnd (email (U? 0.()7 0.02 .0.25 .0,04 0.03 0.11 0.20 0.06 0.39 0.06 0.09 I1 551 i3.23) O 74) (-1 43) (-2.59) (0.27) (0.70) (0.11) (4.22) (0.35) (0.49) (0.02) Oilseeds ()71 4.79 -0.12 .0.07 0.15 0.42 4 0.42 -2.94 2.27 4 0.79 4.30 3.04 (1.23) (490) ( 191) (427) (-190) (3.26) (3.24) (-2.67) (14.15) (-1.19) (-442) (0.83) C-.ugsr I.3 .0.62 0.99 .0.23 .0.31 *1.13 -4.44 15.10 .3.05 -3.59 0.54 .36.41 (I 74) (4,911 (2.24) (4-.44 (1.38) (-1.81) (-657) (1.98) (-2.70) (4.7) (1.13) (-1.47) Fertilizers 0.60 0.02 0.01 .0.83 .0.14 0.35 0.12 .1.44 0.23 0.39 0.05 4.47 (4.43) (0.27) (0.24) (-413) (-.3,4 (3.72) (1.69) (1.116) (2.04) (1.92) (1.00) (1.79) Pesticides 0.61 0.23 0.09 .0.93 -1.01 1.01 0.34 .2.66 0.76 1.20 0.06 10.67 (2.59) (1.90) (1.3S) (3.34) (8.47) (6.48) (2.89) (2.31) (4.57) (1.69) (0.06) (2U) Entergy -2.93 0.12 .0,20 3.75 1.12 -1.62 (.3.97) (447) (-1.031) (4.14) (5.59) (-2.84) Now symmetry and homoxenitry maintaimd. but convexity rot impoed. T-wores, iven tIn ptodhieses, am baed an asympwdg standlard erm.r are given in Table 2.3 A large number of the elasticities are statistically significant and many conform to a pnon expectations. Contrary to conventional iheory, however, the model yields a supply curve for oilseeds which slopes downward in its own price. Reparametenzing the model using the Cholesky-factorization method documented in Lau (1978) and re- estimating the reparameterized model mevealed a single significantly negative Cholesky coefficient. The Cholesky Table 3: Estimated Cholesky parameters. Constrained Unconstrained Parameter Estimates t-scores Estimates t-scores 8, 0.43 0.80 0.88 1.68 822 -()0.16 -2.09 8,1 0.11 1.57 0.09 1.31 8,, 0.69 2.43 0.40 1.91 85,; -0.00 -0.39 0.17 6.38 Note: LR-statistic from the test that 822=0 against the unconstrained estimated equals 13.40. 'Most estimated parameters throughout the paper have been expressed in terms of mean-elasticities; the underlying estimates are available by request. X The ff. 1J I . aN#li shi g ree,i r,ags Larson, et. al. coefficients anid the associated t-scores are given in the first two columns of Table 3. Quasi-convexity of the underlying profit function requires that every Cholesky coefficient be non-negative. Setting the offending coefficient to zero and re-estimating the model a third time produced the second set of elasticities reported in Table 4. Once the second C holesky coefficient has been set to zero the remaining coefficients ar all positive. However, the likelihood ratio statistic resulting from testing the hypothesis that 822-0 against the maintained hypothesis of an unconstrained value for 822, is significantly different from zero.' Therefore, the theoretical assertion of convexity in the profit function is rejected by the data. Table 4: Elasticity estimates at the mean point from estimated normalized quadratic rest eicwd profit model under convexity assumption. with repect to the price of witb tspect to tb qfidty of Quota Ot*r Elaticiry of Cereias Oliaeeds C-auger Pertiliuun Pedcides Ewrgy -Sugr Capital Tim lapus Subsdis Lad Cereals 0.40 .0.04 0.01 .0.21 .0.06 0.10 0.04 . I .0.13 .0.11 004 .0.71 (4.09) (-2.041 (0.641 (-2.88) (.3.10) (-i. 10 (0.23) (1.36) (473) (4.18) (0.41) (.JO) Orliseds -).46 0.05 .0o0 0.24 0.06 0.12 .0.04 -3.62 2.54 .0.56 .0.21 11.83 ,.2 041 098i (40.63) (153) (1.41) (1.65) (433) (-261) (12.11) (466) (-2.35) (2.61) (8 4)n 06 I (o 0.06 .0.17 -1.41 -4.25 15.64 -3.00 -3.92 0.59 -31.70 0 64 i ( 63) (717) (0 06) (4168 (-2.31) (-74) (2.22) (-2.85) (491) (1.29) (-1.42) Femil.ize 4. 0 05 0.00 -0.60 .0.13 1.16 0.08 .1.19 0.20 0.77 0.04 3.78 (2.8ss (I 53) (0L06 (-3.05) (-2.69) (2.94) (1.12) (-1.50) (1.71) (1.61) (0.78) (1.47) Pesncides 090 9 0 In .05 083 1.04 2.72 0.19 2.31 0,67 1.06 0.02 7.61 I. Io) I 'I ((68) )2 ,69i ,-8.65 (4.21) (1.61) (-174) (3.46) (1.30) (0.25) (1.79) 1Energt , IQ 0 04 .009 2.76 1.08 -4.10 ( 76) , 0 59 1 0 (A) (2.98) (4.68) (-.00) Noe symmetr and horngeneit maintained and conventvy impoed. T-scores. given in parentheses, ae baend on uympwtic standrd error.n Setting aside for the moment the implications of rejecting convexity in the underlying profit function, the model results are otherwise quite appealing. The summary statistics of the restricted model are given in the third and fourth columns of Table 1. The statistics are comparable to the earlier results and the goodness-of-fit as measured by the R2 remains essentially unchanged. Most of the mean-point elasticities reported in Table 4 are of the correct sign and significant. As a result of the convexity restriction, all own-price elasticities are of the correct sign and tcur of the six are statistically highly significant. Gtains, oilseeds and fertilizer use are price-inelastic, while non-quota C-sugar exhibits a unitary price elasticity, and energy and pesticide use are price elastic. Ignoring cross- effects, the table suggests that an across-the-board cut in EC support prices in West Gefmany would not affect production output of the major crops (cereals and oilseeds) in the short-run, but would have a larger effect on variable inputs such as pesticides and energy. 'The likelihood ratio statistic is asymptotically distnbuted as a Cti-Square variable with degrees of freedom equal to the number of constraints (Gal!ant and Holly [19801 or Spanos [19861.t In this case the number of constraints equals 1. Since the LR-statistic = 13.40 while X'0,,, = 6.635, the constrained model is rejected. U T'hi ele, ts of '.'i- O5is e Ifrse 'au.i ZPr.in ri al Jointness in production Inherent in the model is the assumption of jointness in crop production. Behaviorly, the assumption implies that vaanous pr(otuction activities are not independent operations, but rather production and mput decisions concerning one output influence the production of other outputs l'his can nse from joint economies (land-rotation is an example) or fTron shared fixed resources. The assumption is intuitively appealing since, in the aggregate, much of the fami-land in West Germany is suitable for a number of crops. From a policy analysis perspective, jointuess in crop production his important implications since policies specifically targeting one crop will have direct spill-over effects for other crops. From a modeling perspective, jointness in production complicates the practice of recovering elasticities by adding a large number of cross-terms, thereby reducing the degrees of freedom for a fixed sample of data. Econometrically, the implication is for less-efficient estimators, especially in the presence of multicollinear.ty. In the context of the present model, jointness in production has two implications. The first is the general notion that unrestnicted supplies are independent of the output prices and supply-managed outputs of other crops, Nested in this general notion is the specific Table 5: Likelihood ratio statistics for non-jointness tests. claim that the supplies of grains, oiLseeds, and non- quota sugar are independent of quota sugar production. Critical The first hypothesis can be tested against the Test LR Statistics XO maintained hypotheses of the base-model by restricting nine of the cross-elasticities to zero. Testing the second notion requires restricting three of the Nonjointness 35.95 21.66 parameters to 7ero. The results given in Table 5 indicate that both altemative models of nonjointness Nonjointness for 29.39 11.34 can be rejected with a high degree of confidence. sugar quota The strong indication of jointness in West German agricultural production implies that quotas on sugar prodluctioti not only create the direct inefficiencies associated with sugar production, but cause distortions in the grains and oiLseed markets as well. Tiese distortions come in the fomi of less-than-optimal production levels, and secondarily. through misallocation of variable inputs as well Measures indicating the extent of these secondary distortions are presented later. Dynamic production It is the static optimization problem which gives nse to many of the properties of the restricted profit function as well as tthe denrved supply and input-deniand schedules. For a given state of available technology and fixed inputs. the famier is hypothesized to optimize his profits for a single period. While the mathematical translation of' propo sed economic activity is somewhat stylized, this standard assumption is perhaps most applicable to the West German farmer growing annual crops who knows with limited uncertainty the price he will receive for his produce and who is free to adjust his crop-mix at the beginning of every season. The implication of static-optimization is that decisions this year are independent of last year's decisions, exclusive of net changes in fixed assets such as capital. This assumption can be tested directly from the data by adding a vector of lagged-endogenous variables to the model and testing the sigrnficance of the addition. Table 6 presents the results of such a test. The static-independence hypothesis is rejected with a high level of confidence Unfortunately. while the results of the single-penr(i optimnization penod Aould suggest that lagged-endogenous variables should not be significant. the significance of the lagged variables does not. in itself. imply any specific alternative theory. Epstein I191). tChambers tI 102. and Chanibers and Lopez (1Q984! have derined dynamic altematives to the static model but empincal applications have been quite limited. 'The inclusion of dynamic elenients in agncultural production currently remains an ad hoc pr(ocedure I( TX ffeI.T of N,hShshmr, 4r,'e ',tev Larson of af Table 6 Test statistics on sigruticance of lagged endogenous vanables 'I'he significance of pricing policies Before proceedmig to the ettecLs ot priinvg and Lagged quota policies, it is perhaps hest to ask zi miore general F3,idogenous Parameter *:-uestion Do prices matter in the short-run.I The * Vanable Estimate t-score existence of short-run price etfects arr generally taken as an article ot taith among ecoriomists. howe er. the fervor of belief is not always shared by policy makers. (irans *().24 -0.92 In the context of the model, the significance of shori- ()ilseeds 0.72 -4.43 run pnce effects can he tested by constraining the C-sugar 0.47 51.8 parameters on all variable inputs and outputs to zero. Penicides 0.47 1.08 The summary statistics resulting from a model which Pesticides 0.47 13.08 binds the pnce parameters to zero are given in Table 7, along with results from a model used to test the Note: the LR-statistic testing the hypothesis that all opposing extreme hypothesis that only prices matter. five parameters equal zero is 110.69; since The LR-statistics, for tests of the hypotheses against the X$00=15.086 the hypothesis is rejcted. maintained assumptions of the base-model are given in __ a note to the table. Surprisingly, dropping the price variables on all variable inputs and outputs from the model has a negligible effect on the summary statistics. The adjusted R's drop slightly and the DW statistic iumproves for the oilseed equation. The same is not true when the fixed supply and input variables are dropped from the equation. The explanatory power drops significantly for all equations. The DW-statistics deteriorate as well. However, as can be seen in Table 8, the original model performs significantly better than either of the altemative models. Both the hypothesis that prices do not matter and the hypothesis that only prices matter can be Table 7. Summar) statistics for estimated normalized quadratic restricted profit function under the assumption that prices do not matter and the altemative that only prices matter. Dependent Adjusted Durbin- Adjusted Durbin- Variable R2 Watson R' Watson Prices-do-not-matter model Only-prices-matter model Grains 0.83 2.22 0.57 0.87 OiLseeds (.98 2.21 0.09 0.19 C-Sugar 0.80 1.67 0.55 1.47 Fertilizer 0.92 1.77 0.77 0.59 Pesticides 0.92 1.12 0.67 0.37 Note: Both altemative hypotheses, that the coefficients on all pnce variables equal zero, and altematively, that the coefficients on all non-price variables equal zero, were rejected with a high degree of confidence. The LR- statistic based on the test that all price-coefficients equal zero was 67.62 compared to the critical value X'`,0=30.58. while the LR-statistic associated with the altemative only-pnce-matters hypothesis was 199.7 compared to a critical value. X",,,=50.892. II The effecu of abC~luhsng green rates Lars5n, ,t 0l rejected with high degrees of confidence. Measures of bias in resource allocation In a multiple output model of agricultural production direct indicators of resource misallocation can be recovered from the estimated parameters. Following Lau (1978a), Weaver (1983), and Moschimu U1988) define the indirect Hicks' neutrality as the following condition: 9) }. - o ° (12) wbere (i,m) represent any pair of variable netputs, and li. is the sth variable quantity (s-i,m) with respect to the fixed factor. Note that when indirect Hicks' neutrality holds the ratio of unconstrained production or input-demand quantity choices is unaffected by fixed inputs or supply quotas. Also, the same condition can be applied to the technology variable to measure biasing effects in technology. Defining the bias measure: -~ *8 (13) T'be constraining level of £ does not bias the mix between two netputs when B",, = 0. When B"1, > 0, the constraint biases the ratio in favor of netput i and against netput m; when B",, < 0, the constraint generates a bias against netput i and in favor of netput m. Table 8 provides the estimated pair-wise measures of bias derived from the indirect Hicks' neutrality condition. The asymptotic t-scores are reported in the table as well. Perhaps surprisingly, a large number are highly significant. Changes in the quota levels for sugar would substantially alter the production mix away from C-sugar and in favor of oilseeds and cereals. The ratios of C-sugar production relative to pesticide and fertilizer demand would decline as well. These results seem logical as an increase in the quota would result primarily in a shift of C-production into quota-sugax production rather than a general increase in sugar production of both types. Changes in the quota allocation would have little effect on the mix between oilseeds and cereals. A decline in existing capital would cause a decline in C-sugar production relative to cereals and oilseeds, as well as a decline in cereals relative to oilseeds. Changes in technology at the mean-point generates a bias in favor of oilseeds over cereala as well as a greater use of fertilizers and pesticides relative to cereal output. A general increase in land availability would appear to favor oilseeds primarily. The biases generated by net subsidies (subsidies minus taxes) are not large but do generate a significant bias in favor of sugar production vis-a-vis cereal and oilseed production and also generate a small bias in favor of cereals over oilseeds. The effects of abolishing green rates In order to quantify the effects of eliminating the policy-determined difference between market and agricultural green rates, six years (crop years 1980/81-1985/86) of EC crop production was simulated under two scenarios. Under the baseline scenario, prices were kept at historic levels. tinder the second, policy prices were reproportioned to reflect market exchange rates. This assumption is no doubt extreme, and it is more likely that a policy which eliminates green rates will be accompanied by either off-setting direct payments to farmers or a general upward revision of policy prices. However simulations of the extreme case re-inforce the general conclusion that short-run production-effects resulting from the policy will be quantitatively small. Table 9 provides the policy exchange rates for rapeseed and cereals, the market exchange rates as well as the changes in policy prices implied by setting green rates equal to market rates for the period under consideration. Under the second scenano the effects of the changes in policy prices on farm prices were assumed to translate in 12 The rffels qf abolsituig greea rages LarMO,. to a1. Table 8: Estimated poir-wise measures of bias for constraining variables at mean-poinx Nerpul Susu-quota Cptal Tecbnology Ohr lepum LAW Net Subsida Pau c qmate "rCco q tmdaIe t.cae caedoaee I4-Wtm mau tocaseta" I-tcom mtab $.*mm (efyal.aAM"Iie 0.07S 0.410 4.500 3.018 *2.671 -10.106 0.450 0.443 .1.538 -2.783 0.238 1.9S6 Ce,ralWK.Suger 4.289 7.237 .14.457 *2.076 2.S61 2.796 3.306 0.906 30.964 1.386 .0.542 -1.255 (emaifiertnilizers -0.041 *0.229 2.374 1.9S6 .0.334 -1.527 -0.50 -1.101 .4493 .1.707 0.006 0.053 CeteaillPa4iles 4.152 40.779 3.491 2.20i .0.7 -3.063 *1.161 .1.141 J5.323 *1.929 0.04 06130 Oll eed(Cnls -0.078 -Q410 -S40 .3.011 2.671 10I006 0450 40.443 12.38 M73 ..25 -1.956 OrileedaC-Suger 4.211 6.575 -19.257 -2.666 5.532 5.153 3.357 0.746 43.522 1.S96 0.799 1.721 Oilaeds/Ferthzt, IS 1 -0.829 -2.426 -1.526 2.337 9.S49 -1.330 *1.372 5.044 .I6 ..250 -2.471 Otii#rdiPtstdclds .0.230 -1.141 .1.306 .0S350 1.S64 5.552 -1.617 .1.170 4.215 0.15 4.232 -1.623 C-Sug.r/Cemyals *4.289 .7.237 14.457 2.076 -2LS61 *2.796 -3.06 4.906 30."4 .1.116 0542 1 C-Sugarlliseea 4.211 46.575 19.257 2.614 -5.532 -5.153 *3.357 4766 *43.322 *1.39 0.7" 1.721 C-Sugsrd1ki1iznr -4.330 -085 16.531 2.335 .3.195 *3.047 4636 -1.097 -35.478 *1.593 0. 1.23 C-Sugr/Pestcides A4.441 -7.293 17.94S L634 -3.668 .3.614 41974 120L4 -39307 1.827 05 1.290 Fertiltizenaereal 0.041 0.229 .2.374 -1.9S6 0.334 1.527 Q.S0 1.101 4.493 1.707 4006 4.053 PFrtatuersA)ilmeds 0.118 0.829 2.426 1.526 .2.337 4.S49 1.330 1.372 .8.044 41.56 0.0 2.471 Fe,tlizrzm/C.Sugar 4.330 6,558 -16S.31 -2.3S8 3.195 3.047 4.686 1.097 35.478 1.543 4.549 -1.206 Fertilize.sAPe,4cides 4.112 -1.395 1.117 1.233 -0.472 -3.604 4U08 40.502 .3.829 -1.341 0.017 0.315 Pesncidr f/ereals 0.152 0.779 -3.491 -2.206 0.507 3.063 1.168 1.141 8.323 1.929 4.024 .0.180 Pe ncid ie" eds 0.230 1.141 1.3t)t 0.530 -1.S64 -5.552 1.617 1.170 -4.215 0.578 0.232 1.62 Peancidn/dCSupr 4.441 7.293 -17.948 -2.634 3.666 3.614 4.974 1.204 39.307 1.827 .4X6 1.290 Pesacidrnfler4iizeru 0.112 1.395 -1.117 *1.233 0.472 3.604 0.28 0502 3.829 1.341 4.017 4.315 T-scoms arc but on uaympko* su.dad enon Table 9: Effects of abolishing green rates on West German policy prices percentage change in Crop green rate market effective policy prices Year rapeseed cereals rate rapeseed cereals C-sugar - ---------- DM/ECU --------------- ------------ % change ------------ 1980/81 2.752 2.752 2.518 -9.3 -9.3 0.0 1981/82 2.657 2.657 2.434 -9.2 -9.2 0.0 1982/83 2.575 2.575 2.315 -11.3 -11.3 0.0 1983/84 2.515 2.528 2.252 -11.7 -12.3 0.0 1984/85 2.450 2.453 2.231 -9.8 -9.9 0.0 1985/86 2.385 2.398 2.169 -10.0 -10.5 0.0 Source: Herlihy et. al. (1989) the following way. Since C-sugar must be exported at international prices, eliminating green rates would have no direct effect on prices received by fafmers. Since green rates are applied directly to intervention prices for cereals, eliminating green rates would reduce cereal prices by the full 9- 10% given in Table 10. Since green rames are applied only to the crushing-subsidy portion of the rapeseed pnce, roughly 50% of the change in the policy price would be 13 The effe ts of aJlussI g ir,een ratts Larson, et. al. passed on to farmers. Th! results of the earlier sections create a dilemma when choosing the appropnate model to simulate the policy chaiges. The hypotheses that prices matter in the short-run and that West German agriculture exhibits jointness-in-production are supported by the data, and all estimated models reported earlier explain a large portion of the deviation in the data. At the same time, the data did not support the hypothesis of quasi-convexity in the underlying restricted profit function, nor did it suppont the insignuficance of lagged dependent variables implicit in a static-optimization problem. Under the working assumption that it is generally best to impose theory on the data, the tesults of the quasi- convex restricted profit function model are reported in Table 11. Hedging all bets, simulation results from the model without convexity restriction as well as the dynamic version of the model are reported as well. Two general conclusions emerge across all three simulations. The first is that substantial reductions in Table 10: Simulated annual percentage changes in selected variables under no-green-rate scenario. Mean Stdv. of Minimum Maximum Change Change Change Change ------ Convexity imposed ----------- Supplies of cereals -1.70 0.25 -2.04 -1.39 oilseeds 1.80 0.58 1.01 2.59 C-sugar -3.13 1.24 -5.51 -2.08 Input demand for fertilizer -2.20 0.33 -2.69 -1.81 pesticides -3.30 0.59 -4.25 -2.58 ---------- Convexity not imposed ---------- Supplies of cereals -4.88 0.13 -1.07 -0.71 oiLseeds -1.59 0.53 -2.38 -0.88 C-sugar -5.60 2.31 -10.10 -3.77 Input demand for fertilizers -2.89 0.44 -3.56 -2.38 pesticides -2.81 0.53 -3.66 -2.20 --------------- Dynamic model ------------- Supplies of cereals -0.56 0.09 -0.70 .0.46 oilseeds -0.25 0.11 -0.44 -0.13 C-sugar 7.79 4.49 4.71 16.22 Input demand for Fertilizers -3.23 0.59 -4.12 -2.46 Pesticides -4.31 1.26 -5.92 -2.44 support prices through a reduction of green rates wil. have a quantitatively negligible impact on output for crops. 14 The eff.t ij of aN14sshoig green rates Lars,o, ft al Recalling that the models fit very well, explainung 80.-90%7 of the deviations in the underlying data, the simulated differenc-s ui supplies and input demand are still within a reasonable range of model error, ranging from 0 to 7%. The result stems from the low price elasticity for cereal and from the uneven way in which green rates affect the three crops modeled. The pnce changes caused by eliminating green rates are only partialy passed on to oilseed prices (via the crushing subsidy). non-quota C-sugar prices are unaffected, and while cereal producer prices receive the full impact of the policy revision, cereal supplies are inelastic in the short-run. The second resul; consistently reported across all three scenarios is that eliminating green rates, with the consequential reduction in cereal and oilseed producer prices, will lead to a reduction in fertilizer and pesticide applications as well as application rates. To the extent that fertilizer runoff and pesticide use generate negative externalities, eliminating West German green rates will result in positive environmental gains producing effects beyond the normal consumer and producer welfare changes. The three simulations offer conflicting results as to the relative changes among the crops. Under the convex-static model simulation, cereal production declines, while substitution effects dominate in oilseeds, leading to a small increase in production. In the unrestricted version of the static model, production of all three crops decline. The dynamic-version of the unrestricted model, price cuts in oilseeds and cereals lead to reduction in the production of those crops and a substitution of productive resources into C-sugar. Concluding remarks The empirical work presented in this paper indicates that the short-run production effects of eliminating green rates on supply and input-demand for cereals, oilseeds, and sugar in West Germany would be relatively small. Despite some savings from reduced input applications, the net effect will therefore be a reduction in farm income. The price effects and the resulting income effects will be disproportately distributed among producers, conversely reflecting the disproportioned benefits of the current system. Producers of cereals have the most to lose by the change. At the same time the analysis convincingly supports the notion that crop production in the EC is joint and that policies aimed at one sector of agriculture have created secondary re. ults in other markets. The quota for sugar perhaps best exemplifies how a policy aimed at one crop in agriculture spills over into production decisions for other crops. In addition, other policy interventions, such as tax-code provisions and direct subsidies are shown to create distortionary effects as well. Policy interventions which have remained in place over a number of years distort the accumulation of fixed resources which have lasting effects. The empirical results indicate that distortions generated by the inappropriate accumulationl of capital generate biases as well. Generally speaking, the results show that the immediate gains in efficiency resulting from supply changes 7re quite limited relative to the immediate costs in terms of price and income reductions faced by West German farmers--despite quite substantive indications of resource misallocations. Therefore, the prospects of long-term efficiency gains must motivate policy aecision-makers to undergo a difficult period of adjustment in the short-run. 15 The efects of aboljAihig gre,, rate Larson, et al APPENDIX A. The Effects of a Currency Realignment on Relative Prices Consider two crops, rapeseed and wheat, in the UK facing a wrrency devaluation of 20%, where: Pre-devaluation: Wheat target price 357.7 ECU Rapeseed target price 464.1 ECU Representative world price 183.5 ECU Rapeseed production aid 280.6 ECU Converting prices to the national currency: Green rate: 0.61865 Market rate: 0.66899 Rapeseed world price (183.5 * 0.66899) 122.73 Rapeseed production aid (280.6 * 0.61865) 127.23 Rapeseed target price in national currency 249.96 Wheat target price in national currency: (357.7 * 0.61865) 221.29 Post-devaluation Rapeseed world price (183.5 * 0.80279) 147.32 Rapeseed production aid (280.6 * 0.61865) 127.23 Rapeseed target price in national currency 274.55 Wheat target price in national currency: (357.7 * 0.61865) 221.29 Conversion of world prices for rapeseed at market rates of exchange allow the full impact of the devaluation to be translated into the national currency. Production of rapeseed has become more attractive (due to higher output prices, expressed in the national currency) relative to wheat. Source: CAP Monitor 16 The uffectJ of aWhAijAg green atetE Lar,on, off di APPENDIX B. Calculation of Monetary Differential Amounts 1. Rapeseed Target prices and published rates of aid (ECU/lOOkg) on 02/15/85 Target Price Aid World Price 50.38 11.04 39.340 2. Exchange Rates for U.K. and West Germany The U.K. green pound is worth more than the market rate of the pound and the German gren mark is worth less than the market rate against the ECU. UK green rate: 1 ECU = itO.618655 or 'iY = 1.61641 ECU UK agricultural market rate: Yl = 1.58691 ECU - 1.033651 (CRCF-coefficient) therefore Sil = 1.53525 ECU West German green rate I ECU = DM 2.38516 or DM I = 0.41926 ECU West German agricultural market rate DM I = 0.446062 ECU + 1.033651 (CRCF-coefficient) therefore DM I = 0.43154 ECU 3. Calculation of Rapeseed Subsidy in national currency without MDA adiustment UK West Germany i. Target price (50.38*green . e) 5[31.17 DM 120.16 ii. Aid (11.04*green rate) St 6.83 DM 26.33 iii. Net cost in national currency V24.34 DM 93.83 iv. Net cost in ECU 37.39 ECU 40.49 ECU v. World price 39.34 ECU Without MDA adjustment, UK rapeseed is cheaper to UK crushers than is West German rapeseed to West Germany crushers. The net cost in the UK is below the world .rice so the subsidy is too high and vice versa for West Germany. The coefficient represents a central rate correctinR factor. Starting in the marketing year 1984/85, for each product, a coefficient (central rate correcting factor), is applied in agrimonetary calculations, including MCA/MDAs. This is equivalent to revaluing the ECU for agricultural purposes and cuts positive MCAs at the expense of increasing negative MCAs. The central rate correcting factor is adjusted following EMS realignments. 17 Th offr(r5 , akh)5hjng gi ten rates Larson, et. al. 4. Calculauon of the basic MDA percentages MDA percentages e(qual the percentage divergence of green rates from agricultural market rates: UK MDA% = I - (1.61640/1.53525) * 100% = -5.286% German MDA% = I - (0.41926/0.43154) * 100% = 2.846% If the MDA differs by less than one percentage point from the existing MDA, the existing MDA continues to apply. 5. Ap2lication of MDAs to current (spot) rates of aid The basic MDA percentage is applied to both the target price (first element) and to unadjusted rate of aid (second element). If the MDA is positive, it is positive on the target price and negative on the aid; if the MDA is negative, then the reverse, UK West Germaoy Target price 231,17 DM 120.16 Aid 6.83 26.33 MDA first element - 1.65 3.42 MDA second element 0.36 - 0.75 MDA adjusted aid 5.54 29.00 Net cost in national currency 25.63 91.16 Net cost in ECU converted at agricultural market rates 39.34 ECU 39.34 ECU Where the seed is crushed in another member state other than the one in which it was harvested, the rate of aid is converted using the rates published in the Official Journal. If UK rapeseed was crushed in Germany, the aid in DM would be: UK adjusted aid is: Y5.54 - bilateral Y/ECU rate: 0.618334 * bilateral DM/ECU rate: 2.22732 = DM 19.95 Note that this is less than the aid in DM for West German produced rapeseed because the green rate support system currently means that UK rapeseed prices are lower than West German. Source: CAP Monitor 18 The eftfrs 'f1 abolislung green rates Larson, v a' Bibliography Amernmiya, T., 1983. "Nonlinear Regression Models.' Hand-Book of Econometrics, vol 1, ed. Z. Griliches and M. D. Intriligator. Amsterdam: North-Holland Publishing Co Gallant, A. Ronald, and Alberto Holly, 1980. "Statistical Inference in an Implicit, Nonlinear, Simultaneous Equation Model in the Context of Maximum Likelihood Estimation", Econometrica 48, pp 697- 720. Herliby, Michael, Stephen Magiera, Richard Henry, and Kenneth Bailey, 1989. Agricultural Statistics of the European Community, 1960-85. Economic Research Service, Statistical Bulletin No. 770, United States Department of Agriculture. Washington, DC. Josling, Tim. August, 1989. "Europe 1992 and CAP Reform", Notes for presentation at AAEA syr'posium, unpublished. Kelcb, David. July, 1989. "Europe 1992: Implications for Agriculture" in Western Europe: Agriculture and Trade Report. Economic Research Service, RS-89-2, United States Department of Agriculture, Washington, DC. Lau, L. J., 1986. "A Charactenzation of the Normalized Restricted Profit Function." Journal of Economic Theory 12, pp 131-63. ,1978a. "Applications of Profit Functions." Production Economks: A Dual Approach to Theory and Applications, Vol 1, ed. M. Fuss and D. McFadden. Amsterdam: North-Holland Publishing Co. , 1978b. "Testing and Imposing Monotonicity, Convexity, and Quasi-Convexity Constraints." Production Economics: A Dual Approach to Theory and Applications, Vol 1, ed. M. Fuss and D. McFadden. Amsterdam: North-Holland Publishing Co. Lee, H. and R. G. Chambers, 1986. "Expenditure Constraints and Profit Maximization in U.S. Agriculture." American Journal of Agricultural Economics 68, pp. 857-65. Lopez, E. R., 1984. "Estimating Substitution and Expansion Effects Using a Profit Function Framework." American Journal of Agricultural Economics 66, pp. 358-67. McFadden, D., 1978. "Cost, Revenue, and Profit Functions." Production Economics: A Dual Approach to Theory and Applications, Vol 1, ed. M. Fuss and D. McFadden. Amsterdam: North-Holland Publishing Co. Moschini, G., 1988. "A Model of Production with Supply Manr gement for the Canadian Agricultural Sector." American Journal of Agricultural Economics, pp. 318-329. Pfouts, R. W., 1961. "The Theory of Cost and Production in the Multiproduct Firm." Econometrics 29, pp. 650-58. Ray, S. C., 1982. "A Translog Cost Function Analysis of U.S. Agriculture, 1939-77." American Journal of Agricultural Economics 64, pp. 490-98. 19 The effecu (yf abofxiung green rutes LaFrsonl, et. al. Sakai, Y., 1974 "Substitution and Expansion Effects in Production Theory: The Case of Joint Production." Journal of Economic Theory 9, pp. 255-74 Shumway, C. R., 1983. "Supply, Demand, and Technology in a Muluproduct Industry: Texas Field Crops." American Journal of Agricultural Economics 65, pp. 748-60. , R. D. Pope, and E. X. Nash, i984. "Allocatable Fixed Inputs and Jointness in Agricultural Production: Implications for Economic Modeling." American Journal of Agricutural Economics 66, pp. 72-78. Spanos, Aris, 1986. Statistical Foundations of Econometric Modelling, Cambridge: Cambridge University Press, Swinbank, Alan. September, 1988. "Green Money, MCAs, and the Green ECU: Policy Contortions in the 1980s" in Agra Europe Special Report No. 47, Agra Europe, London. Weaver, R. D., 1983. "Multiple Input, Multiple Output Production Choices, and Technology in the U.S. Wheat Region." American Journal of Agricultural Economics 65, pp. 45-56. 20 FIRE Worjk-a Pap' Series Contact ~~~~~~~~~~~~~Ail, ... Aor 12& to, paper WS"j')i.; T >z.; F '; pi 1 d . ....t ,'' > Brent B3orrel l i ebruiry 199¶ P Kokia Pr')h V !J L')'QKt! 33,:'16 W FS`9,' R1 '!t S!',l Ig 't'ie M I F ,r e Ref:k Erzan February 1991 G logon Arranrg,;rnerit Theory and Eviderice Kala Krshna 33732 from U S A;)pa.el Imports frorTm Ling Hui Tan Hong Kong WPS598 Africa Region Population Projections: Patience W. Stephens February 1991 0. Nadora 1990 91 Edt!ion Eduard Bos 31091 My 1. Vu Rodolfo A. Bulatao WPS599 Asia Region Population Prolections Eduard Bos February 1991 0. Nadora 19S0-91 Edition Patience W. Stephens 31091 My T. Vu Rodolfo A. Bulatao WPS600 Latin America and the Caribbean My T. Vu February 1991 0. Nadora Region Population Projections Eduard Bos 31091 1990C91 Edition Patience W. Stephens Rodolfo A. Bu'atao WPS601 Eurooe, Middle East, and Nornh Eduard Bos February 1991 0. Nadora Africa Region Population Pro;ections Patience W. Stephens 31091 1990-91 Ed,tion My T Vu Rodolto A Bulatao NPS602 Firm Output Ad;ustment to Trade Mark A Dutz February 1991 S. Fallon L beralization Theory w th 38009 Appiication to the Moroccan Experience WPS603 The Role of Of iciaHly Supported Ash Dernirgi6q-Kunt Febiuary 1991 G. llogon Expol Cred t5 in Sub-Saharan Refik Erzan 33732 Africa's External Financ ng WPS604 Foreign Trade and Its Relation to Faezeh Foroutan February 1991 S. Fallon Competition and Productivity in 37942 Turkish Industry WPS605 Overview of Contractual Savings Dimltri Vittas March 1991 W. Pitayato- Savings Institution: Michael Skully nakarn 37666 WPS606 Adjustment Policies and Investment Luis Serven March 1991 E. Khine Performance in Developing Andres Solimano 39361 Countres: Theory, Country Experiences, and Policy Implications F'lHF Vio'.Kz1 Paper Se,ej .k':, A A C H I' '. 1)c .i aT l jr-n M-arr' 19c i 1) GAtalsa E rr'' rB r ., Jo, £ 'co e-r jr WPSG©5 Crass Co,:r,!y Stuc' as cf Grow?', Ross, L evie Mlarco 1991 C[E:CMG ad~l Poiicy Metnoao;ogca Dav.d Rene 1 391 75 Ccrcotn,and S!la¶stcal P'obla-> WPIS§C9 A Sos:vyAraws s~ a' C'oss Ross LeVnre Miarch 1991 CECMG Court-y GroWth Reg'essions David Renait 391 75