ifdp · March 31, 1984

Oil Prices, Welfare Tradeoffs, and International Policy Coordination in a Three Region World Model: An Optimizing Approach

International Finance Discussion Papers Number 241

April 1984

OIL PRICES, WELFARE TRADEOFFS, AND INTERNATIONAL POLICY COORDINATION IN A THREE REGION WORLD MODEL: AN OPTIMIZING APPROACH

Jaime Marquez and Peter Pauly

NOTE: International Finance Discussion Papers are preliminary materials circulated to stimulate discussion and critical comment. References in publications to International Finance Discussion Papers (otner than an acknowledgment by a writer that he has had access to unpublished materials) should be cleared with the author or authors.

Title: Oil Prices, Welfare Tradeoffs, and International Policy Courdination: An Uptimizing Approach

Authors: Jaime Maryuez and Peter Pauly

Address: Board of Governors of The Federal Reserve System Washington, D.C. 20551, and Economics Department, University of Pennsylvania Philadelphia, PA. 19104

Date: April 1984

Series and Number: International Finance Discussion Papers, No. 241

Pages: 59

Price: No charge

JEL Subject Classification: 212,713

Keywords: Oil prices, Optimal control, Policy coordination, Welfare

The first purpose of this paper is to derive optimal oil prices by recognizing that oil price changes affect real income of oil importers and vice versa. International trade flows, income, and prices are endogenously determined in a three region world model (VCs, UPEC, Won-UPLC LbCs). Usiny Chow's method for stochastic control of a general nonlinear dynamic system, we then offer a characterization of the set of efficient cooperative policy strategies for the country blocks considered here, given a set of weights in the welfare function. We also examine the sensitivity of the optimal policy mix to changes in the

weights of the welfare function.

OIL PRICES, WELFARE TRADEOFFS, AND INTERNATIONAL POLICY COORDINATION IN A THREE REGION WORLD MODEL: AN OPTIMIZING APPROACH *

Jaime Marquez International Finance Division Federal Reserve Board and Peter Pauly Economics Department University of Pennsylvania April 1984 Abstract. The first purpose of this paper is to derive optimal oil prices by recognizing that oi] price changes affect real income of oi] importers and vice versa. International trade flows, income, and prices are endogenously determined in a three region world model (DCs, OPEC, Non-OPEC LDCs). Using Chow's method for stochastic control of a general nonlinear dynamic system, we then offer a characterization of the set of efficient cooperative policy strategies for the country blocks considered here, given

a set weights in the welfare function. We also examine the sensitivity of the optimal policy mix to changes in the weights of the welfare function.

I. INTRODUCTION

Our first purpose in this paper jis to endogenize OPEC's pricing policy recognizing that real income of oil importers is affected by oi] price changes and that oil price changes are affected by real income of oi] jmporters. Our second purpose here is to ascertain the existence of efficient cooperative policies among developed countries, OPEC, and Non-OPEC developing countries, and the extent to which alternative policies result in welfare tradeoffs for these country blocks.

In endogenizing oi] prices, we relax two of the most common assumptions found in the analyses of oi] price effects and of oil price . determination, namely that oil price changes are exogenously given "shocks", and that the oi] demand that OPEC faces is either a stable function or one

2 that shifts at a known (exogenous) growth rate. Price taking behavior is

an assumption that may be valid in the short-run, or for countries with very little participation in the oi] market. However, for countries like the U.S., Japan, and West Germany, with oi] consumption representing a large share of the international oi] market, the assumption of oi] price exogeneity is not tenable. An increase in oi] prices affects their real income, their oi] consumption, and therefore feedback to OPEC's oil revenues and oi] prices.

As a corollary of removing the small country assumption, we relax the assumption of stability of the oil demand curve faced by OPEC, which is the typical assumption made in analyses of oil price determination.

Assuming a stable demand curve implies that of] price changes have no effect on either the level of real income or its growth rate, which is in contrast to the “supply shock" literature and to ten years of empirical findings. Thus we can see that--up to now--the effects of oil price changes and the determination of oi] prices have been analyzed in a dichotomous manner since what has been assumed as exogenous in one kind of analysis is treated as endogenous in the other and vice versa (see Fischer et. al., 1975).

In this paper we remove this dichotomy by treating oi] prices and economic activity as jointly determined. This joint determination of oi] prices and real income implies that OPEC faces, in addition to the problem of allocating oi] production through time, a tradeoff between exploiting the direct price inelasticity of oil demand on the one hand, and avoiding the income feedback effects of oil price increases on the other. In particular, if the price path is too low, then OPEC does not exploit the price inelasticity of oil demand and thus incurs revenue losses. However, if the price path is too high, then real income of oil importers is adversely affected and this feeds back to OPEC in the form of lower oi]

exports and oi] revenue losses.

It would seem that once we allow for the fact that oi] price

changes, and the associated policy responses, have implications for the

world economy, the issue of international policy coordination among the

country blocks we consider is a natural question to address. A good deal of

attention has been given to the possibility of sub-optimal equilibrium

3 | Tevels with "low" levels of economic activity resulting from uncoordinated

macropolicies. However, the analyses have been characterized by

deterministic, static, two-country models, with little or no empirical

content (an exception is Oudiz and Sachs 1984). This paper also addresses

the question of international policy coordination aimed at restoring a non-

inflationary recovery of the world economy. But in doing so, we use a

stochastic, dynamic, three-country model, which allows us also to

empirically address other interesting questions:

What are the optimal fiscal and monetary policy responses in industrial economies to changes in oi] prices?

What is the degree of substitutability among various policy instruments? Can lower oil price changes be a substitute for capital transfers to LDCs in acheiving a given growth target in developing countries? Can these capital transfers be avoided by coordinated expansionary fiscal policy in DCs?

What is the importance of income feedback effects in determining the price of oil from OPEC's perspective?

What is the nature and sensitivity of welfare tradeoffs among DCs, OPEC, and Non-OPEC LDCs to different coordinated policy mixes?

How different is the oil] price path that maximizes welfare for oi] exporters from the oil price path that maximizes welfare for oi]

importers?

Several conclusions can be drawn from our analysis. First, recognizing the income feedback effects of oi] price changes raises the absolute value of the demand price elasticity and reduces the oil price path consistent with oi] exporters’ best interests. Second, the international transmission of oi] price shocks generates changes in the distribution of income between regions; therefore, alternative coordinated policies, as developed in this analysis, have different effects on the prospects for the major regions of the world.

We begin the analysis in section II where we develop a small econometric, three-region world model to study the international transmission of oi] price effects on real income, prices and international trade, assuming that oi] price changes are exogenous. In section III we endogenize oi] prices by applying optimal control to the econometric model to formally derive a system where oi] prices--along with other exogenous variables--and real income of oi] importers are simultaneously determined. We implement our approach empirically in section IV, and section V contains

our conclusions.

II. A GLOBAL MODEL OF OIL PRICE EFFECTS

The model we use is simple but comprehensive and it is based on theoretical models developed by Taylor (1981) and Marquez (1983). We divide the world economy into three regions: (1) the developed economies, whose GDP js determined from the demand side; the supply side is introduced via a production function which determines prices and factor input demands; (2) the OPEC countries, for whom we only explain the recycling of oi] export

earnings to purchase imports of manufactures, and (3) the Non-OPEC

5 developing countries, whose GDP is determined from the supply side using a production function. Centrally planned economies have been excluded from the present model.

The international trade flows included in the model. are: (1) oil, sold by OPEC to both developed and Non-OPEC developing economies; the latter alsc export oi] to developed economies; (2) manufactures, exported by the DCs to both OPEC and Non-OPEC LDCs; and (3) manufactures and raw materials, which are exported by the Non-OPEC LDCs to DCs. In a compact form, the model has 52 equations, 16 of which are behavioral. The parameters of the model are econometrically estimated using data for 1960-1979, although some of the relationships are estimated using data only up to 1977. The estimation method we use is OLS since the advantage of alternative parameter estimators such as 2SLS, 3SLS, and FIML need not hold for small samples such as ours (Mariano 1978). We now describe the main behavioral equations which

are shown, in general form, in Table 1.

Developed Economies Equation (1) represents private consumption, cf, depending on a

distributed lag of nominal value added PY ya deflated by the consumption

price index Poon} the short run mpc is 0.47 and the long run income elasticity is one. Investment, 1 in equation (2), is a function of a distributed lag of both real income and the long term nominal interest rate rd, the short run mpi is 0.13 and the long run elasticity with respect to income is 1.18; the long run elasticity with respect to the interest rate is -0.20.

Imports of raw materials from Non-OPEC developing countries, ne in

Table 1

Main Behavioral Relations in

Global Econometric Model

Developed Economies

OPEC

= #(pd yd, pd = (Py vay Pe

on)

F(y4, 4)

f(P./ pd, yA)

= F(ARP,, ASP y, ABPQ 1, Ur) f(L,[K,E(0,C)J)

F(Pys Pos Pps Pps GY9)

04. of

=d “n & d M, - X2(Po/ P., G9)

=( Po/ Py )( Md + XE )

Ma= (Po (Md+ me)/pa )

Non-OPEC Developing Countries Gy*= #(K% 0%)

pay

ie te

F(PE/ Py, YE R4/ Pd) F(Ix, R4/ Pa)

f(Pay pe, yd)

f( PE/P, , GY*)

2 ASR, yk 0°- 08% + x2

(1) (2) (3) (4) (5) (6) (7) (8) (9)

(10)

(11) (12) (13) (14) (15)

(16)

7 equation (3), are a function of (i) present and past values of the price of

raw materials, Pas relative to the export price of manufactures of DCs, pd

and (ii) real income of DCs, y7, the long run relative price elasticity is -0.69 and the long run income elasticity is 0.63. The inflation rate of the export price of manufactures, ape in

j equation (4), depends on (i) the rate of change of nominal wages, AwP J with a coefficient of 0.327; (ii) the inflation rate of oil prices, ABP)» lagged One period with a coefficient of 0.09; (iii) the inflation rate of raw materials' prices with a coefficient of 0.34; and (iv) a measure of capacity utilization, U, obtained as the difference between potential output and actual output.

Potential output, Y*, is estimated as a trend of actual output, but we split the period of estimation into two subperiods: 1960-1972 and 1973-1979. One reason for such a sample split is to allow for the possibility of structural changes that might have taken place after the first oi] shock in 1973, such as the slowdown of productivity growth. The estimated growth rate for potential output for the period prior to 1973 is 4.7% and 2.9% for the period afterwards. A Chow test for the null hypothesis of structural parameter stability could not be accepted.

The demand for oi] (in barrels), of in equation (6), is a conditional demand function derived from a three level CES production function, equation (5), whose arguments are labor, L, capital, K, ofl, 0, and coal, C. As a result, both capital-energy substitution possibilities and interfuel substitution possibilities are taken into account. The conditional demand for oil depends on the prices of oi1, coal, labor, and

the rental price of capital as well as on gross output (value added plus oi]

imports), cy?, We estimate a linearized version of this demand function with a distributed lag (4 periods) for prices and income, allowing for homogeneity of degree zero in prices. In estimating this relationship, we use switching regression with a split of the sample in 1972. A Chow test for the null hypothesis of parameter stability could not be accepted. The long run oil price elasticity declines from -0.27 to -0.57. The long run coal price elasticity increases from 0.29 to 0.91. The estimated income elasticity declines from 1.70 to 1.34. The hypothesis of homogeneity of degree zero in prices cannot be rejected.

DCs' total imports of oi] (in barrels), ac in equation (7), are equal to their demand for oi] (in barrels) minus their exogenously given supply of oil, o® (also in barrels). Imports of oi] from OPEC, ne ‘in equation (8), are equal to total imports of oi] minus imports of oi] from Non-OPEC LDCs, % » which in turn depend on the price of oi] relative to the price of coal as well as on gross output of DCs. Since the product of total

oil imports in real terms, we

O° and the oil price index, Po? is identical to

the product of total imports of oi] in barrels, Ac, and the price of oi] PS (in dollars per barrel), we can solve from this identity for oi] imports in

real terms as in equation (9).

OPEC

The only behavioral relation we explain is OPEC's absorption capacity of oi] revenues. For this we relate their imports of manufactures, M in equation (10), to a distributed lag of oi] revenues deflated by the export price of manufactures of DCs. The absorption elasticity in the first

year is 0.31, 0.36 in the second year, and 0.04 after four years. The long

run absorption elasticity is 0.98.

Non-OPEC Developing Countries

Gross output (value added plus oi] imports in real terms), Gy” in equation (11), is determined using a two level nested CES production function with capital, K*, and oi] consumption, 0%, as arguments. The parameters of this function are estimated in two steps by using the firstorder conditions for cost minimization and the production function itself. The short run elasticity of substitution between oi] and capital is 0.05 and the long run elasticity is 0.73.

Following Coen (1971), we determine capital formation, 1% in equation (12), as a function of the rental price of capital, Pre relative to the price of oi], and on gross output. We model the speed of adjustment of investment as a function of the availability of foreign exchange resources in real terms, Rey / pe . We find an inelastic response of investment to changes in both income and relative prices. However, we find investment to be quite responsive to changes in foreign exchange reserves (an elasticity

in excess of one).

Imports of manufactures, Me in equation (13), are derived using the fact that the capital stock is an aggregate of the domestic capital stock and the foreign capital stock, with a positive (but finite) elasticity of substitution. Following Marquez (1984), we obtain imports of manufactures as a function of net investment, I, and foreign exchange reserves. Our results point to an elasticity of imports of manufactures with respect to investment of 0.8 and an elasticity with respect to (real)

foreign exchange reserves in excess of one.

Exports of manufactures to DCs, xt in equation (14), depend on (i) the export price of manufactures of DCs relative to the export price of manufactures of LDCs, pope, and (ii) the real GDP of DCs. 0141 consumption, o* in equation (15), is modeled as a conditional demand function derived from the production function, and it depends on relative factor prices and gross output, cy*. Using the identity between world consumption (in barrels) and world oi] production (in barrels), we derive LDCs’ imports of oi] from OPEC (in barrels), My » as the difference between LDCs' oi] demand and (exogenously given) oi] supply, os, plus exports of oil] of LDCs to DCs. Finally, we link oi] imports in barrels to oi] imports in real value following the same steps as in equation (9) for DCs.

For the purposes of section III, it is particularly useful to express the econometric model of Table 1 in its state-space representation

as:

¥e= F(yy» Yt-1 Xt> Xteepe Zer 8) +e > (17)

where Ye= vector of endogenous variables, f = vector of functions,

= vector of controllable exogenous variables,

z= vector of non-controllable exogenous variables, 6 = vector of parameters,

e = vector of random errors, e ~ N(0,Q),

Q = estimated covariance matrix.

The state-space representation is obtained by redefining both

endogenous and controllable variables with lags of order greater than one

11 as new variables. In addition, we have classified exogenous variables as

either instruments, X42 OF non-controllables, 24.

III. SIMULTANEOUS DETERMINATION OF OIL PRICES AND REAL INCOME: AN OPTIMIZING APPROACH

The optimizing approach, as developed originally by Niehans (1968), contains two key ingredients. The first one is a social welfare function providing a ranking of the alternative combinations of target values. The second ingredient is the existence of a target frontier specifying the maximum value that a given target can obtain for given values of the other policy targets. The application of this optimizing approach to our problem has several advantages. First, there exists a quantitative criterion to evaluate systematically the effects of alternative instruments ' paths on the behavior of the economy. Secondly, the "principle" of effective market classification is embedded in the optimizing approach. By this principle, instruments are assigned to that target on which it has relatively the strongest effect. This is an important point because the solution to an optimal control problem takes the form of an efficient "policy mix" allowing for the relative desirability of targets, the relative costs of changing the instruments, and the dynamic behavior of the economy.

An alternative to the optimizing approach is a "trial-and-error" approach where one examines the effects on the economy of a large number of alternative paths for the exogenous variables. However, and in contrast to optimal control, there is no objective criterion to choose an efficient policy mix among the alternative paths for the exogenous variables. In order to select the best policy mix for the controlling

variables, we minimize the expected cost of not achieving a given set of

12 income growth targets for each of the regions considered here, subject to the behavioral constraints and identities embodied in the econometric model

of section IT:

T * * min E(W) = Ef I (yp> ag) 'K (ye a4) t

where % = (Ye: x,)3 K is a positive definite diagonal matrix of weights; a, is a vector of desired values for targets--such as OPEC's oi] revenue growth--as well as instruments--such as oi] price changes. The choice of a quadratic welfare function is for mathematical convenience. |

Stating the optimal control problem in this form is equivalent to maximizing the weighted sum of each regions' welfare function. Given an appropriate choice of weights (the K matrix), and under the assumption of cooperative behavior among policymakers, it can be shown (Niehans 1968, Oudiz and Sachs 1984) that the solution to the first-order conditions of the

above optimal control problem (the entire path for x,) is the efficient

) policy choice. A policy choice is said to be efficient if there exists no other policy vector capable of reducing one country's welfare losses without increasing welfare losses to another country. Alternatively, a policy mix is said to be efficient if the associated target values belong to the target frontier. The solution of the first-order conditions for the above

minimization problem involves applying Pontryagin's minimum principle (see Marquez 1983), and it takes the form of a linear feedback rule linking real

income, and other endogenous variables included in » to oil prices and Xt-1

other exogenous variables included in xy!

. * X= G48 aes K) Vey > (18)

Notice that G. is a time varying matrix with elements depending on the (estimated) parameters of the model, the desired values for instruments and targets, and the elements of the K matrix. In using a feedback rule we recognize that (1) economic policies are not set once and for all, i.e@., they are revised (with a lag of one period) as information about the future values of the endogenous variables becomes known, and (2) current policies have an impact on future policies. This is a relevant issue since a change in oi] prices today will have a lasting influence on oi] consumption and this must be taken into account when determining the price of oil.

Having derived an expression for the optimal policy path, equation (18), we combine it with the econometric model to obtain an "enlarged" econometric model where real income growth, an element of Yeo and (optimal)

oi] prices, an element of Xe are jointly determined:

Yt = fly» Yt-1° Xe Xt-1° 2° 8) + et > (17)

x4 = G,(85 Ke a Yea (18) To illustrate this joint determination, suppose we are given an initial path for oil price's growth, x Jy Using equation (17), we determine the effect that the given path of oi] price changes will have on real income of oi] importers. These income effects, together with the substitution effects, are transmitted to OPEC in the form of both movements along and shifts of the oi] demand schedule. Using equation (18), these changes in the oi]

demand schedule are now taken into account by OPEC (with a lag of one

14 period), to obtain a second oi] price path {x }ywhich, by equation (17), is used to recompute the effects of oi] price changes on economic activity. This second round of oi] price effects is, in turn, used to update the previous oil price path and so on. This iterative process between oi]

prices, real income, and oi] demand continues until OC h- oem 10,

where s is the sth iteration.

IV. EXPERIMENTAL DESIGN Description of the Welfare Function In order to implement empirically our approach, we parametrize the

welfare function as :

min E(W) =E{ : Wee (Vea= ac, )° + ci : kas (Xag= Pan) ?} = t=1982 j=] Jt Jt jth” where the target variables are: Yi the growth rate for real oi] revenues of OPEC, ARR?: Y5= the growth rate for real income of developed economies, avy9; Y3= the growth rate for Non-OPEC developing countries, any’; and Yq= the inflation rate of the export price of manufactures of DCs, agp" The policy instruments we use are: Xi the growth rate for oil prices, ARP 3 Xo= Tong term nominal interest rates of developed economies, rd, X3= real government expenditures in DCs, 67; and X,= nominal net capital transfers to Non-OPEC developing countries, CF.

In postulating such a welfare function we have assumed that OPEC,

in order to accelerate their development process, sets the price of oi] in order to secure a flow of oi] export earnings large enough to finance the foreign component of their capital stock (Aperjis 1982, Marquez 1983). We also include in the welfare function growth rates for developed and Non-OPEC developing economies to study both changes in the distribution of income between DCs, OPEC, and Non-OPEC LDCs and the feasibility of a high growth path for the world economy. Finally, the inflation rate of the export price of manufactures, a proxy for the overall inflation rate, is included to capture policy makers' concern with allocative inefficiencies and distributional effects of high and uncertain inflation rates (Niehans 1968, Marquez and Vining 1984).

With respect to the instruments, we use the growth rate for oi] prices since it is the instrument that OPEC uses to ensure a steady flow of export earnings. We include government expenditures and interest rates of developed economies because they allow us to study what type of coordinated fiscal, and (only indirectly) monetary policies can be implemented by developed economies in order to achieve a non-inflationary growth path for the world economy. Finally, capital transfers are included to study the financial requirements implied by sustained growth in the Locs:

By considering a number of policy instruments, we can obtain an jsotarget curve by examining the extent to which instruments can substitute ~ for each other in achieving a specific target. For instance, a particular growth path for Non-OPEC LDCs could be achieved with alternative paths for oi] prices and government expenditures in DCs, and one issue that arises here is to what extent lower oil prices can be a substitute for an

expansionary fiscal policy in DCs given the LDCs' growth path.

Parameters of the Welfare Function

The desired values for targets and instruments are shown in Table 2. We assume a target growth rate for developed economies of 2.5 percent in 1982, which is steadily increased to 5.8 percent in 1989 and 1990. For developing countries we assume a target growth rate of four percent in 1982, which is steadily increased to seven percent for the period 1987-1990. Real oi] revenues of OPEC, R°, are targeted to grow at one percent during 1982 and 1983. Although these growth rates are small relative to the standards of the 1970's, they are ambitious in relation to the performance of real oil revenues for the 1982-1983 period. We increase this growth rate from four percent in 1984 to ten percent in 1989 and 1990. The target inflation rate of exports of manufactures is seven percent in 1982, steadily declines to five percent jin 1985, and remains at five percent for the remainder of the horizon.

Nominal oi] prices are targeted to have zero growth rate in 1982, a decline of 15 percent in 1983, and a six percent increase for the 1984- 1990 period implying a target of one percent increase in real oil prices for this last period. Long term nominal interest rates, rd are targeted to decline from 14 percent in 1982 to eight percent by 1986 and to remain at this level for the remainder of the horizon. For government expenditures we assume an average annual growth rate of only 0.9 percent, reflecting the concern over increasing government deficits of current administrations. Capital transfers to LDCs in nominal terms are assumed to grow at an 8.5 percent growth rate per year starting in 1980.° Government expenditures and capital transfers enter in the welfare function as indices with 1982=1.0.

Our first characterization of the efficient policy mix consists of

a baseline optimal control solution with an egalitarian welfare function,

Table 2

Base Optimal Control Solution Desired Values for Targets and Instruments (1982-1990)

Targets

agy? ASR° asy” aepe 1982 2.5 1.0 4.0 7.0 1983 3.4 1.0 4.8 6.5 1984 4.0° 4.0 5.5 5.5 1985 4.3 6.0 6.0 5.0 1986 4.8 7.0 6.5 .. 5.0 1987 5.2 8.0 7.0 5.0 1988 5.5 9.0 7.0 5.0 1989 5.8 10.0 7.0 5.0 1990 5.6 10.0 7.0 5.0 Instruments

AgP rd Gd CF

fo)

1982 0.0 14 1.0 1.0 1983 -15.0 11 1.01 1.11 1984 6.5 10 1.02 1.23 1985 6 10 1.03 1.37 1986 6 g 1.04 1.52 1987 6 8 1.05 1.70 1986 6 8 1.06 1.87 1989 6 8 1.07 2.U8 1990 6 8

1.08 2.31

AMN-22 CHO NO TAETOBRO

eaNaDe aamMNamaZz—

Figure 1

Mean Fitted Values for Instruments: Base Optimal Control Solution

Woy

we a

- ~

DUSUPPUTy TUVUUUTONN LOVTRYUETN TUT TUUEOTY [UVEVOTETY IVETTYVWT YUVUVYYOTL [VEVEVETT TOTTTTTT

t: desired values a: fitted values

a ~

8 3 3 3

Figure 1

(continued)

. w o

e °

2 ~g fs 8

SOrwa FTE wKewFo—e Dew

teteii restate i... oo lito oa let tit losaees bles heed Beek Bee OOes DRO |

TIME

ee - 5 ee 2 mw

oo OY) - - - - - ~ -

Odeeray Ket EMUWeEN —~O YOU w

TIME

Figure 2 Mean Fitted Values for Targets:

Base Optimal Control Solution

desired values ‘fitted values

jeerecenenscerone

ee taaahs bedhead beeeee Seeeel ROAAne BOO SOd DODOOE RORDSE SORE DROOad LOOROS |

RHF vae-—O2R 2ar-wW Ov KaAFDUGUr DewWMH

6 w

2 wo

~ w

o 2

TIME

JPR Rp erperrerren prennen peerrrrrsp rr — e » ” _ 7” e ~ ' @a2o0z2-F On AV 4

8s 86 87 68 89 TIME

u ©

Figure 2.

(continued)

in aa o e

« Ty Dr re ye Pr eer eer

wo * w oe w oe ww oe mw eo e “ cv) o yy ” Ls o° « w : = = 2

wwruw7 DU ErOBH FE Ou CAWe

83 eu 8s 86 87 88 6a ge TIME

SH I pe

“ s s a

Eerot- rT Cu 400

“

epee

” OY]

e cy

8s 86 ‘87 88 39 3e TIME

22 1.e@., we assign a weight equal to one to each target variable. The tradeoffs between targets that may arise here are due to the behavioral constraints embedded in our model and to the values given to the instruments' weights. The iterative algorithm used (Closed Loop) is sensitive to the choice of instruments' weights, especially low values (similar findings are reported by Klein and Su 1980). We select the values for the instruments’ weights taking into account that policy changes are not likely to be large or frequent (Klein 1983). Thus, we give a large weight to both real government expenditures of DCs (99999) and nominal net capital transfers to LDCs (1500). Interest rates are assigned a weight of 20 and

Oil prices are assigned a weight of 4.5.

Empirical Results

We apply the optimal control algorithm developed by Chow (1975, 1981) to the econometric model of section II for the period 1982-1990 given the targets and weights described above. Figures 1 and 2 contain the mean fitted values for both the instruments and targets.

Starting with the instruments, we notice that the fitted values for the interest rates are very close to their desired values. Nominal oi] prices increase by 3.2 percent in 1982, decline by 8.8 percent in 1983, and increase in 1984 by seven percent; for the 1985-1990 period nominal oi] price changes remain in the six percent range. Finally, mean fitted values for both real government expenditures and nominal capital transfers are very close to their targeted values.

With respect to the targets, arp? decreases from 5.9 percent in

1982 to 3.9 percent in 1984, but rises again to 5.8 percent by 1990. The

initial decline in anps is due to both the fall in oi] prices in 1983 and the reduction in economic activity of DCs in 1982. Growth in developed economies, after experiencing a decline in activity in 1982 of one percent, shows a steady increase from 3.2 percent in 1983 to 4.5 percent in 1990. OPEC's real of] revenues decline sharply in 1982 and 1983 (16 percent in each year) because of both the recession in of] importing countries and the decline in oi] prices. However, these revenues increase by 10.8 percent in 1984, 6.5 percent in 1985, and maintain an annual average growth rate of 5.8 percent for the period 1986-1990. Growth in Non-OPEC developing countries is positive for the entire planning period starting with a three percent growth rate in 1982, with a small decline to 2.1 percent in 1984, and increasing steadily to 5.6 percent in 1990; the average annual growth rate is 3.5 percent.

Recognizing both the limitations of our model, and given the particular weights used, we find that if international policy coordination results in (1) nominal oi] prices growing at an average of six percent per year, (2) nominal capital transfers to Non-OPEC developing countries growing at 8.5 percent per year, (3) long term interest rates declining to 8.0 percent, and (4) real government expenditures continuing their (modest) expansionary path, then it is feasible to achieve relatively high growth rates for developed, OPEC, and Non-OPEC developing countries for the

remainder of this decade.

The Importance of Income Feedback Effects and OPEC's Pricing Policy

One of the main contentions of this paper is that allowance for the income feedback effects of oil price changes raises the absolute value

of the price elasticity of oi] demand and therefore lowers the optimal price

24 path consistent with OPEC's best interests. What is not so clear, however, js whether these feedback effects are important enough to result in significantly different oi] price paths.

One way to evaluate the importance of these feedback effects is to contrast the effects on oi] revenue growth of a price path where oj] prices grow at a rate equal to the interest rate (Hotelling 1931, Stiglitz 1976) and thus no feedback effects are allowed, against the effects on oi] revenue growth of an optimal oil price path derived using optimal control, where feedback effects are allowed. Table 3 contains our results, which are depicted in Figure 3.

For every year, except the first two, the growth path for OPEC's revenue in the no-feedback case is below the revenue growth path derived using optimal control, even though oil prices grow faster in the no-feedback case than in the optimal control case. The deterioration of OPEC's revenue growth in the no-feedback case occurs because the revenue increase arising out of short-term price inelasticity of of] demand is offset by the reduction in real income of oi] importers, which dampens oi] consumption and shifts to the left the export demand schedule faced by OPEC. In other words, the income feedback effect¢ of oil price increases raises the (absolute value of) the price elasticity above 1 and therefore further price increases lead to revenue losses. And what our results imply is that it is optimal to increase the price of oi] until the revenue increase arising out of short-run price inelasticity is equal to the revenue loss arising out of

the income feedback effects.

Sensitivity Analysis

The characterization of the efficient policy mix derived above is

conditioned on both a particular weighting matrix and on the assumption of

1982 1983 1984 1985

1986

1987 .

1988

1989

1990

Table 3

Optimal Vil Prices With and Without Feeaback Effects Allowed

No Feedback Effects

S&P

14 ll 10 10

Allowed

A&R

ASY

Feedback Effects Allowed

ASP

A&R

fe)

AtY

pemo[ Te sJIVjJja Yoeqpeesz ou (q PeMoTTe sqosjye Yoeqpee;s (e

HLAOND SNNAASN S$ Badd § 9d 30 S194 HiMOs) JOTdd WO A0 SLVe HLMON ora OF @ Ol- O&- g. oo @ Gh~

oS — _ cr)

fy 06 ' Oo ; > oe ! ( 48 \ ag a f 88 ‘ PR od i 1 ‘ 28 , ; 98 | ‘i }! : 1 d it } G8 fy f ‘ ! £ \ 3 1 8 Te os vB a " . ‘Sore. ore aoe oon, “Tokes £3 S57 £3 ‘ f _— H : mae f a g Vv Gt 6 “

$}029j3J3T AORQPI9T JnoyIfM pue YITM S2°FId TFO TewFado

€ sansty

cooperation among the three blocks of countries in setting the instruments' values. Therefore, it is of interest to perform a sensitivity analysis to study the robustness of our results to changes in the values of the weights. In particular, we analyze the welfare tradeoffs among OPEC, developed countries and Non-OPEC developing countries that arise as a result of changes in (1) the preference ranking of targets, and (2) the cost attached to fluctuations in the instruments, particularly oil price changes, real government expenditures, and nominal capital transfers to LDCs.

One useful definition of welfare losses, W., is:

1990 We = (1/9 J) (Ya, ~ a5,))1/%, §-1,2,3. t=1982

W. measures the average deviation of the fitted value of variable Y; from its targeted value. Therefore, the smaller is W.5 the smaller is the "welfare" loss of the ith country block. For instance, the last row of Table 4 contains the welfare losses for the three regions considered here, under the baseline optimal control solution. Our results indicate that, on the average, the DCs' growth rate deviates 1.7 percent from the target growth rate. OPEC's revenue growth shows the greatest deviation, 8.7 percent, which could be due to either the setting of extremely

ambitious targets or to the particular choice of weights. We concentrate in the remainder of this paper on the sensitivity of these welfare losses to

changes only in the weights of the welfare function.

Case I: Greater Weight to OPEC's Revenue Growth Target We now study the welfare tradeoffs between the three country

blocks that arise when the weight for OPEC's revenue target is increased

\ t

Weights

po g

kK K 4.5 99999 2.3 99999 O.1 99999 4.5 5000 2.3 5000 0.1 5000

Base Solution

Table 4

Welfare Tradeoffs Case I: Greater Weight tor UPEC'S Revenue Growth Target

Welfare Losses

DC's

OPEC

@ Averaye growth rate, 1982-1990.

LDC's

Ww N e ° oO o

Ww ° N

6.4 8.3 20.6

29 from one to ten. A welfare tradeoff js said to occur when a decline in W. jnduces an increase in some other Wie We also examine the sensitivity of such tradeoffs to different weights for oil price changes (0.1, 2.3, 4.5) and for real government expenditures (99999, 5000). Table 4 presents the welfare losses for each country block and the average growth rate for oi] prices corresponding to alternative combinations of weights; Figure 4 depicts the results of Table 4. Figures 5 and 6 illustrate the behavior of targets and instruments for selected weight values.

We begin the sensitivity analysis by increasing the preference given to OPEC's growth target while leaving unchanged all the remaining weights. We find that OPEC's welfare losses are reduced from an average of 8.7 percent for the base solution to 4.8 percent in the present case. This reduction #s possible because of the average growth rate of oi] prices increases from 4.3 percent to 8.1 percent. At the same time, the reduction in OPEC's welfare losses induces an increase in the welfare losses of both developed and developing economies by 0.6 percent and 0.1 percent respectively. Thus if the OPEC countries have enhanced bargaining power, then they may be able to reduce their welfare losses only by inducing an increase in welfare losses to the other countries, which shows that the income distribution for the original set of weights was efficient.

As mentioned earlier, one of the advantages of using optimal control is that the instruments are assigned to targets according to their relative efficacy. This assignment, however, depends on the relative costs attached to fluctuations in the different instruments, and for this reason we examine how sensitive is this assignment to changes in the instruments’ relative weights. Suppose then that we reduce the value of the weight for

changes in oil] prices. Intuitively, the smaller the value of the weight for

a3d0 Ld) s.907 KY 5.30 01 $SO7 3805734 §331ad 110 404d LHOL3H

eees

I aSvVO SLHDIGM LNAWNULSNI NI SA9NVHO OL SASSOT FUVATAM AO ALIALLISNAS

y BAnsTy

Figure 5

31 Mean Fitted Values for Instruments: Case I

e a t

K =99999 i)

8 K =5000

=2.3, fo] =0.1, fo] 8a

desired K P K P 87

Oe re ne ee

t 66 TIME

86 TIME

es 8S

63 8.

r e 2 = e e os . ss bl 2 bb x 2 w > m” nu = = bd ”

. : . ‘ hy 2 VCS “Y 5 on ., ‘ \ os N . * N 4 YON NY . ‘ + \ ‘, N ‘ a. 4 \ ‘ o f H o ‘ \ H : \ $ , \ H i 3 : ; . \ 4 ‘ \ } \ ‘ \ $ t ‘ s hi Ii ; . i i : ii 7 and ] . { hag ] . / : / / / H . ccrrayversnenepneneetjsnerscnayenerr anager npr net ever eyepreeevenrey speveevenen pre en een e peeve reens pre eveedeeprenrrerer [evs rerees peeverevesprvereeg pT eTeeTe g 7 n Oo - ow w e 5 2 5 : j ; — Be Wor Lae aareuw

OxOR- TF OW OM BAKOWY

Figure 5

(continued)

syerveveceeperereeyryyrrrrrs res | revererysprry

wn ~ Co] -

OOK WE FEU HAW O—> DOW

1.088

. . . Ly . ry * . . * *

eeverepreissrerepersscnrscgercrercespencvererey!

TIME

2» ~

- oe a oe & o w

OGRE Hau &-O

7 on

100 7

“

nat heed Aaah hahah baked DAOae heed heed Dane Lheed bende Bheed boned baked hand Bie

TIME

Figure °6

Mean Fitted Values for Targets: Case I

a6 87 as 89 TIME

@, \ N / N N / \ / N N NN \ “ . aN NN NX N \ \ \ \ \ \ \ oO oO Oo oO oO ov oO roa) wy il i] oo od se mw MM s, a a se, oo o+ ed oe . ” HN oO we wl ll u a a oO ° vo a (27) N mM Mm NY y? ee ee oo 7 ra rN) we >, vevewervsjeevecevenqverrrvent(ervevsrpeprenimerurprssrercegerrrerercy, S Wotan, e o e rs . o “ w 7 .

exo ET Oh OF HF Pa SOF Aes OU NARDIN AI DHS

Figure 6

(continued)

t -8

- - on"

A H : : t 8 3 : : . ‘ .

ets en 002"

eae nenesy

‘ eer yprree perc pe rer yr rev eres eye pee Dee hele

a “ o ” oO ” e e mt o e « « o « w i e “ - . ? e - - -

. 7 .

euwr>uraw aeeods-z ou coeur

—

TINE

SWssr eee nnnenne’

™.,

“peer egpeere gy rrr yee ewer prereset en ae ered ee |

oe w co ad e wn e we w os o w wo ” . om

eaeotr-x Cu YEU

w or}

TIME

35 a given instrument the more willing we are to use that instrument to improve OPEC's growth target. In particular, a reduction of the weight for oil price changes from 4.5 to 0.1 has the effect of raising the average growth rate of oi] prices from 8.1 percent to 19.8 percent, which increases welfare losses to both DCs and LDCs by 1.3 and 1.5 percent respectively without reducing OPEC's welfare losses. The increases in welfare losses of DCs and LDCs are not more pronounced because as the average growth rate for oi] prices increases, the optimal fiscal policy response becomes more expansionary and thus tends to offset the adverse effects of increases in 011 prices on income growth.

Fiscal policy in DCs might also be used as an alternative instrument to achieve OPEC's growth target. To examine this possibility we reduce the cost of fluctuations in this instrument from 99999 to 5000. As a result, we find that fiscal policy fluctuates more freely, and that it leads to an unambiguous reduction in welfare losses for each block of countries. This “Pareto” improvement is possible because the optimal response of government expenditures in developed economies is rather expansionary, exceeding its target by an annual average of $US 55 billion, whereas the optimal fiscal policy response with a large weight exceeds its target only by an average of $US 8.6 billion. As a result of this expansionary fiscal policy there is a stimulus to oi] imports from OPEC which, when combined with a greater average growth rate in oil prices, explains the reduction in welfare losses to OPEC. Similarly, we find a substantial reduction in welfare losses to LDCs despite (1) increases in the average growth rate of oil prices, and (2) virtually no deviations of capital transfers from their desired values. This is because higher government expenditures in DCs serve as a substitute instrument for capital transfers to LDCs since it increases

imports of DCs

36 from LDCs and this increases foreign exchange reserves with the same impact

on growth of LDCs as a transfer of capital.

Case II: Greater Weight to DCs Growth Target

In this case we begin by increasing the weight of DCs' growth target from one to ten while leaving all the other targets with a weight of one. The results are shown in Table 5 and are depicted in Figure 7. Figures 8 and 9 illustrate the behavior of targets and instruments for selected weight values.

In general, the initial impact of an increase in the weight for DCs growth target is to raise government expenditures in DCs. However, as we saw in Case I, there is a direct relation between movements in of] price changes and movements in optimal real government expenditures in DCs. This positive relationship implies that the effects of an increase in government expenditures on real income of DCs are offset by the increases in oi] prices that immediately follow. This offsetting effect requires an even larger expansion of government expenditures to counteract the negative effects of higher oi] prices and to achieve the growth target. However, any further increase in government expenditures is followed by another round of of] price increases which again increases government expenditures. The net effect of this process is an increase in welfare losses to DCs and LDCs but a reduction in welfare losses to OPEC. This adverse result on DCs welfare losses arises because the increase in oil prices is of a greater macnitude than the increase in government expenditure given the large weight differential between these two instruments. |

In order to eliminate the influence of a large weight differential on the welfare tradeoffs, we reduce the weight of government expenditures

from 99999 to

Table 5

Welfare Tradeoffs Case II: Greater Weight for DC's Growth Target

Weights Welfare Losses a Koo Ky DC's OPEC LbC's A&P. ce pn 4.5 99999 2.2 7.8 3.1 6.1 2.3 99999 2.3 7.4 3.2 7.1 0.1 99999 2.6 6.6 3.2 9.2 4.5 5000 +~—s3D 8.3 3.1 5.6 2.4 5000 2.0 8.1 3.0 5.1 0.1 5000 1.6 5.0 3.0 10.9

Weight for DC's Growth = 50 4.5 5000 1.3 8.3 2.8 4.7 Weight for DC's Growth = 200 4.5 5000 0.6 7.6 2.5 5.2

Base solution 1.7 8.7 2.7 4.3

nn nn

4 average growth rate, 1982-1990.

2340 [ } s.901 KAWSY s.00 01 SSO1 340473" $301ad 110 404 LHO13K

eees

azey YyAnory AL. a2FId TFO aberaay

66666 sa1nq};puedxy Quawuzaacy 103 AYybyemM

I] ASVO SLHDIAM LNAWNULSNI NI SHONVHO OL SASSOT GaVATaM JO ALIAILISNGAS

L 2an3sty

. 8

Figure

Case II

Mean Fitted Values for Instruments:

as a6 67. 86 ag 3@ TINE

i 8 i, \ oa ¥ a o>) \ a (o>) h e oa °o. I ° Oo Yay Ki Wo Woo I mo OM ii « . Ht: 3 gq «a H oO - . H Hn oO : ae | i} ga 8 : e mo MM mM Ww 0 po) sf 8 / / 1, ! a } s i ! ! ! . at Ceaaataet neti anMMaR MIO OE RE OO rrapererenennyes zh g SaSaa aia? Naaee Sale AO RR RR SR “ory

Ee wer arate OWMOR-E CK OF 4 BAH OWH

Figure 8 (continued)

rerrveres, |. wry [reevreveryreresee ry erevererryyerreerrs jrrerverry greys rveTs prey ere eT yey Terr ef

re :s 8&8 $s £€ $ 8 & §& 8 bad

on - pay = re) a ~ ~ - _ _ - - - e

BOrwe sewer wWHEwEaA—r Few

TIME

. eeoeepetrecettctel tocol bist ood oa] tales aces baded bead baeed Benes aed Od mn - © ee ne © ww © Hm MW =F F&F 2 & > a od

Oae--ar -eaREonwes +O YOU

TIME

9 pate AR Ce SRR AE lag ES MaRS Sth

Figure 9

Mean Fitted Values for Targets: Case II

desired

\ Hy \ 5 \ 4 \ i oO \ ‘ ON (on) \ ‘ an oO \ an oo \ H n ww \ Hl oD ] aD \ 4M \ 4 / H a a 7 no od / i xn oO / ' wow / H

K K

po po

‘e

° e e

eee eee,

TO pr prep rrerporreps rer preppy rep rss

a & 3 e 2 x & e ry os “ “ “ “

Rea OF Saree) Te FAT DU cies eis

66 87 98 as

TIME

vgs 3 Py

TE) roel] oleh belated bade e sees) Dee ee eee Rebbe o wo » ” « - e

OFNB-T on OU wo

TETT ET Bev:

TIME

Figure

(continued)

: 7 * ° a 7 #6?

PRD OPO B- F Ae. OH wed

TIME

7.a<

ae a) “ o

Oe sa Be tenet eee eee ae ee a ee

ROBT On

jou mo

wo ~

43 5000. We find that this weight reduction leads to a decline in welfare losses for DCs from 3.0 percent to 1.6 percent and for OPEC from 8.3 percent to 5.0 percent with almost no effect on LDCs welfare losses despite an increase of the average growth rate for oi] prices from 5.6 percent to 10.9 percent. This last result suggests that the adverse effect of higher oi] prices on both DCs and LDCs growth are offset by the stimulative effects of increases in government expenditures of DCs, given that this instrument can fluctuate more freely now.

We also examine the impacts of additional increases in the weight for the DCs growth target. In order to detect nonlinear intertemporal tradeoffs, we consider an increase in this weight first from 10 to 50 and then from 10 to 200, while keeping the weights for oil prices and government expenditures at 4.5 and 5000 respectively. Our results indicate that when the weight for DCs equals 200, their welfare loss declines from 3 percent to 0.6 percent. There is a small decline in welfare losses of OPEC (0.2 percent) even though the average oj] price growth rate declines under the new weighting scheme. This suggests that OPEC's reduction in welfare losses arises out of the increase in oil] exports induced by the higher growth in developed economies. Finally, Non-OPEC LDCs experience a decline in welfare losses from 3.1 percent to 2.5 percent because the effects of higher government expenditures (higher growth in DCs) and lower oi] price growth are transmitted to them via international trade, enlarging their foreign exchange resources and thus reducing their need for capital transfers above the targeted levels. Thus, once again, we find that increases in DCs' government expenditures serve as a substitute instrument for capital

transfers to Non-OPEC developing countries.

Case III: Greater Weight to LDCs Growth Target

In this case we begin increasing the weight of LDCs target growth rate from one to ten while keeping the weights for the other targets equal to one, We also increase our willingness to use capital tranfers more freely, and thus we reduce the weight for this instrument from 1500 to 100. Our results are shown in Table 6 and depicted in Figure 10. Figures 11 and 12 show the behavior of instruments and targets for selected weight values.

An increase in the weight of LDCs' target growth rate results in an unambiguous reduction in their welfare losses by an average of 1.1 percentage points with respect to the baseline solution for all the combinations of weights considered here. Moreover, we notice that even if we let oil prices fluctuate more freely, by reducing their weight from 4.5 to 0.1, there is no increase in welfare losses to LDCs despite an increase in the annual average growth rate of oi] prices from 4.3 percent to 11.4 percent. This is because the adverse effects of higher oil prices are offset by higher capital transfers, which exceed their targets by an annual average of $US 19.5 billion. Additional effects of the increase in the average growth rate of oi] prices are a reduction in OPEC's welfare losses from 8.6 percent to 5.6 percent and an increase in DCs' welfare losses from 1.6 percent to 2.6 percent.

The effectiveness of using government expenditures to achieve the LDCs growth target depends on the behavior of oi] prices. In particular, if fluctuations in oil price changes have attached a weight of 4.5, then a reduction in the weight on government expenditures does not affect LIiCs at all, and adversely affects DCs and OPEC. This is because with a weight of 4.5, oi] prices tend to follow their target values quite closely, which includes a reduction in the average growth rate for oil prices. The fall in oil prices leads to a decline in government expenditures, given the positive

assocation between these two instruments, which explains the adverse effects

Case III:

___weights K

Table

6 -

Weltare Tradeoffs

Weltare Losses

Greater Weight for LDC's Growth Target

po Kg DC's OPEC LDC's asp,” ace P 4.5 9999Y 1.6 8.6 1.6 4.3 12.1 2.3 99999 1.7 7.8 1.6 5.1 13.4 QO.1- 99999 2.6 5.6 1.6 11.4 19.5 4.5 5000 1.9 8.7 1.6 3.8 14.3 2.3 5000 1.8 7.9 1.5 4.6 14.8 QO.1 5000 2.0 4.8 1.5 11.4 17.7 Base Solution 1.7 2.7 4.3 0.2

® Average growth rate, 1982-1990.

b Average deviation from taryet, 1982-1990, billions of

SUS.

a3ad0 [___} s.903 KS s.00 701 SSO 3uus73N $301ad U0 30s 1HO1 3H

00S

TW adSvo SLIOTMM LNAAMLESNI NI SA9NVHO OL SASSOT FYVa14dM AO ALIALLISNGS

OT ean3sty

ano-a2e —-O NO ZALORO

OM AMMAN EWs

' ~

‘ a

Figure 11

Mean Fitted Values for Instruments: Case III

we \ . t: desired

a: K_ =0.1, K_=99999 po 8

e: K_=0.1, K_=5000 po g

TINE

/ \ \

‘ “ [\Wreweu pevers eurery

(weveed prevarrerr.s

67 8a es 99

48.

Figure 11

(continued)

Perr phe Ce ee oe Bd ee rererecypre = o o 2 in wo a - - = e oe = e E e

VOSWHF TWEE wHawFOrN- SOY we

. Vp yp rey

TIRE

’ rrevyrerryeryry myer yrerepreeryrerrprenyren peetperyprepreeey 8 > 2 e

a o e w 7 ” ~ ~ - - - oo - - ~

Oder a FaeAEOUWAEH -O ABO

~ 7“

au 8s 66 67 96 89 TIME

Figure 12

Mean Fitted Values for Targets: Case III

fey) fon) oO roa) oO fon) oO fon) Ya) Li} i foo} feo} mM MM. a a Or te | oO - ° 4 O oO A i oe oe co MM Mm re vu <[eves reevey vers en ers yreveyy veeyerverenrygues $s Lg & 3 z « o . ” ry

e 3 8 &

“ “ o w

Pa ITE HOF HK -w) OW FAF De. dt Dem

Pe acacia Rebcahchdidehihd Loidtihehhd bibddiddd bedii did

TIME

Sprrereerrege cerns prren rere rye entree pert erry re ee oe o w 2 on” ~ ~ oe

eeos- Fr Ou CU

arenes Tv

TIME

Figure 12

(continued)

_—<—" —_

roe

a “—

AVUD COCO E CET ENT EN TENT TRTUT TED HOTEL UF UE STEEL OTE UTUTVUETYT STE TROTTUI LTT UNI TOTY ‘ t 1

7- wn oe rd oe w oe n w ~ rr 0

euwrwrow OfOB-X Ch OAWL

TIME

_—-—--—- 8 a

-_—"

rtd

baieebtddd Lied eedidd iedbedaed bebbdbeadd babiddhads bedhddbddd LibAdadlhd aaanaehl kaadedanad hl

a “ e ” e ” ° “ - « o e “ w » 2 “ a

eeo8-F Cu 300

as 86 87 se a9 ry) Time

on DCs and OPEC. We also notice that, despite the decline in the average growth rate of oil prices, there is an increase in the amount of capital transfers to LDCs. This increase is necessary to offset the impact that lower growth in DCs has on the LDCs, since the beneficial aspects of lower oil prices are being offset by the reduction in growth of DCs.

In contrast to the previous situation, a reduction in the weight of oil prices to 0.1 means that oil prices can deviate more freely from their target path. As a result, a reduction in the cost of fluctuations in government expenditures from 99999 to 5000 marginally reduces welfare losses of LDCs and significantly reduces the welfare losses of both DCs and OPEC. In addition, we notice that it is possible to have a decline in average capital transfers to LDCs since DCs' government expenditures are now higher, which increases imports from LDCs and enlarges their foreign exchange

receipts.

Evaluation of Sensitivity Analysis Results

Several interesting results emerge from the extensive sensitivity analysis performed so far. First, although instrument instability did not materialize for the weights we have used, there are cases where small weight values result in paths for instruments that are not entirely consistent with the policy environment where these instruments are determined. In other words, even though a policy mix might be efficient, it may lack credibility when compared to historical standards. To be specific, we find that in order to accelerate economic growth in Non-OPEC LDCs, capital transfers would have to exceed the (already high) targets by an annual average of $US 20 billion in some cases, and in no case is this excess below an annual

average of $US 12 billion. Given today's current financial situation for

52 some developing countries, it seems unlikely that such large transfers will take place.

Second, we find that the oi] price path consistent with OPEC's best interests is higher than the oi] price path consistent with either DCs or Non-OPEC LDCs' best interests. This result is robust with respect to the different combinations of weights considered here, as can be seen from Figures 4, 7, and 10 by comparing the average growth rate for oi] prices for each of the weighting schemes we have considered. This finding clearly indicates the existence of conflicting interests among the three blocks of

countries included in our model, and calls for an extension of our analysis

to allow for non-cooperative policy design.

V. CONCLUSIONS

Our first purpose in this paper has been to study the international transmission of oil price effects and the determination of optimal oi] price paths, not as separate problems but rather as one problem by recognizing that economic activity and oi] prices are jointly determined. We then study the set of efficient policy paths for oi] prices, government expenditures, and capital transfers to LDCs consistent with a noninflationary reactivation of the world economy. The effects of changing the weights of the welfare function on the income tradeoffs among different regions and on the optimal policy mix are also examined.

We start our analysis by developing a nonlinear stochastic dynamic econometric model for a three region world economy that highlights the channels of international transmission of oi] price effects and the feedback effects of ofl price changes on the demand for oi]. The simultaneous

relation between oi] prices and real income is obtained by minimizing a

53 quadratic welfare function subject to the econometric model. We apply optimai control to an estimated version of this model and we find that if (1) of] prices grow at six percent per year, (2) government expenditures grow at. 0.9 percent per year, and (3) capital transfers to developing countries grow at 9 percent per year, then it is feasible to achieve a noninflationary recovery of the world economy.

Given the structure of the model, its estimated parameters, the optimizing horizon, and the optimization algorithm we find that:(1) the growth rate of oil prices taking into account income feedback effects is smaller than the growth rate of oil prices when these feedback effects are not taken into account; (2) real oi] revenues of OPEC grow at a faster rate when income feedback effects are allowed jin the determination of oi] prices than in the case where these income effects are not allowed; (3) there is a positive relationship between the optimal response of government expenditures and optimal oil] price changes; (4) low oil prices and expansionary fiscal policy in DCs serve as substitute instruments for net capital transfers to developing countries; (5) there exists a conflict of interests between oil importers and of] exporters since the oi] price path that reduces OPEC's welfare losses is higher than the oi] price path

required to reduce welfare losses for both DCs and LDCs.

54 ENDNOTES

An earlier version of this paper was presented at the 5th IAEE Annual North American Meetings, June 1983, and the 4th IFAC/IFOR conference on The Modeling and Control of National Economies, June 1983. We are grateful to F. Gerard Adams, Matt Canzoneri, David Kendrick, Lawrence Klein, Peter Tinsley, Janice Shack-Marquez, Edward Green, and James Sweeny for their comments. We are also grateful to Gregory Chow for providing the software used in our analysis. Any remaining errors belong entirely to us. This paper represents the views of the authors and should not be interpreted as reflecting the views of the Board of Governors of The Federal Reserve

System or other members of its staff.

See Hamilton (1983), Marion and Svensson (1982), Bruno and Sachs (1979), Findlay and Rodriguez (1977), and Schmid (1976). 2

See Hotelling (1931), Solow (1974), Cremer and Weitzman (1976), Pindyck (1978), and Salant (1982).

See Niehans (1968), Hamada (1979), Johansen (1982), Canzoneri and Gray (1983), and Oudiz and Sachs (1984).

In order to be fair, we should point out that optimal control applications are not problem free. In particular, Kydland and Prescott (1977) argue that policies derived using optimal control need not be optimal if agents have rational expectations. The main problem is, as they indicate, the absence of a mechanism inducing future policymakers to take

jnto account the effects of their

policies on today's agents decisions. However, the issue is far from resolved since Chow (1981) shows how to apply optimal control theory

allowing for rational expectations.

Capital transfers to Non-OPEC LDCs can hardly be considered a variable under their control. A more reasonable interpretation in our context would be to treat them as transfers required from international

cooperation.

Capital transfers here are net of amortization and interest

payments, and thus they represent "new money".

A similar target growth rate is found in the United Nations

Development Report (United Nations 1982).

This growth rate is similar to that of the “high-transfer

scenario" of the World Bank Report for 1982 (World Bank 1982).

The optimal control solutions converged in three iterations. 1 0 e ° e

This growth rate is higher than the "high scenario" target growth rate for developing economies (3.3%) of the 1982 report of the World Bank (World Bank 1982:37). 11 .

It might be possible to argue that the higher revenue growth

derived using optimal control solutions is not due to income feedback

effects, but rather due to the response of some other instrument such as

government expenditures, which stimulates DC's real income and thus OPEC's revenue growth. To make sure that an instruments’ responses, other than oi] prices, are not responsible for the rapid growth in OPEC's revenue, we have given a rather high weight (99999) to government expenditures. We find that the average deviation of government expenditures from their target is, in absolute value, equal to 0.9 percent, which can hardly be considered as a reason for the increase in OPEC's revenue growth.

nf One way of relaxing the assumption of cooperation is, for instance, to include in the welfare function only OPEC's targets and instruments and have reaction functions for DCs and LDCs. A second approach is to have a welfare function for each of the players who have conflicting objectives (Pindyck 1977).

v8 Another instance is the case of greater weight to OPEC's growth target using a low weight for government expenditures. These expenditures exceed their targeted values by an annual average of $55 billion in real terms. Again, given today's concern with government budget deficits, it seems unlikely that such an expansion of government expenditure will

materialize.

57 REFERENCES

Aperjis, D. (1982). The Oil Market in the 1980's: OPEC Oil Policy and Economic Development. Cambridge: Ballinger Publishing Company.

Bruno, M., and J. Sachs (1979). “Macroeconomic Adjustments with Import Price Shocks: Real and Monetary Aspects." The Maurice Falk Institute for Research in Economics. Discussion Paper 793, Jerusalem, Israel.

Canzoneri, M. and J. Gray (1983). "Two Essays on Monetary Theory in an

Interdependent World." International Finance Discussion Paper No.219, Board of Governors of The Federal Reserve System.

Chow, G. (1975). Analysis and Control of Dynamic Systems. New York: John

Wiley. Chow, G (1981). Econometric Analysis by Control Methods. New York: John Wiley.

Coen, R. (1971). "The Effect of Cash Flow in the Speed of Adjustment." In G. Eromm (ed.) Tax Incentives and Capital Spending. Washington, D.C.: Brookings Insitution.

Cremer, J. and M. Weitzman (1976). "OPEC and the Monoply Price of Oil." European Economic Review. Vol. 8, No.2: 155-164.

Findlay, R., and C. A. Rodriguez (1977). "Intermediate Imports and Macro-

economic Policy under Flexible Exchange Rates." Canadian Journal of Economics. Vol. 10, No.2: 208-217.

Fischer, D., D. Gately, and J. Kyle (1975). " The Prospectus for OPEC: A Critical Survey of Models of The World Oi] Market." Journal of Development Economics. Vol.2, No.4: 363-386.

Hamada, K. (1979). " Macroeconomic Strategy and Coordination Under Alternative Exchange Rates." in Dornbusch and Frenkel (eds.)

International Economic Policy: Theory and Evidence. Baltimore: Johns Hopkins University Press.

Hamilton, J. (1983). "Oil and The Macroeconomy since World War II." Journal of Political Economy. Vol. 91,No. 2: 228-248,

Hotelling, H. (1931). "The Economics of Exhaustible Resources." The Journal of Political Economy. Vol. 39, No.2: 137-175.

Johansen, L. (1982). "A Note on the Possibility of an International Equilibrium with Low Levels of Economic Activity." Journal of International Economics. Vol.13 No.3/4: 257-265.

Klein, L. (1983). “Economic Policy Formulation: Theory and Implementation (Applied Econometrics in the Public Sector)." In Handbook of Econometrics, Vol. 2, forthcoming.

Klein, L. and V. Su (1980). "Recent Economic Fluctuations and Stabilization

Policies: An Optimal Control Approach." In L. R. Klein, M. Nerlove, and S. Tsiang (eds.) Quantitative Economics and Development. New York: Academic Press.

Kydland, F. and E. Prescott (1977). “Rules Rather Than Discretion: The Inconsistency of Optimal Plans." Journal of Political Economy. Vol. 85, No.3:473-491.

Mariano, R. (1978). “Commodity Market Modeling: Methodological Issues and Control Theory Applications." In F. G. Adams and J. Behrman

(eds.) Econometric Modeling of World Commodity Policy. Lexington: Lexington Books.

Marion, N. and L. Svensson (1982). "Structural Differences and Macroeconomic Adjustments to Oi] Price Increases in a Three-Country

Model." National Bureau of Economic Research, Working Faper No. 839, Cambridge: National Bureau of Economic Research.

Marquez, J. (1984). "Foreign Exchange Constraints and Growth Possibilities

in LDCs." Journal of Development Economics, forthcomming.

Marquez, J. (1983). "The International Transmission of Oil Price Effects and the Derivation of Optimal Oi1 Prices." Ph.D. dissertation, Philadelphia: University of Pennsylvania.

Marquez, J. and D. Vining (1984). “Inflation and Relative Price Behavior: A Survey of the Literature" in M. Ballabon (ed.) Economic Perspectives Vol. 3: 1-56. New York: Hardwood Academic Publishers.

Niehans, J. (1968). "Monetary and Fiscal Policies in Open Economies Under Fixed Exchange Rates: An Optimizing Approach." Journal of _ Political Economy. Vol.76, No.4: 893-920.

Oudiz, G. and J. Sachs (1984). "Macroeconomic Policy Coordination Among the Industrial Countries." Paper presented at the Brookings Panel on Economic Activity, April 5-6, 1984,

Pindyck, R. (1977). "Optimal Economic Stabilization Policies Under Decentralized and Conflicting Objectives." IEEE Transactions on Automatic Control. Vol. ac-22, No.4: 517-530.

Pindyck, R. (1978). "Gains to Producers from the Cartelization of Exhaustible Resources." Review of Economics and Statistics. Vol. 60, No.2: 238-251.

Salant, S. (1982). Imperfect Competition in The World Oi1 Market, Lexington: Lexington Books.

Schmid, M. (1976). “A Model of Trade in Money, Goods, and Factors." Journal of International Economics. Vol. 6, No.4: 347-362.

Solow, R. (1980). "What to Do (Macroeconomically) When OPEC Comes?" In S. Fischer (ed.) Rational Expectations and Economic Policy.

Chicago: National Bureau of Economic Reserach.

59 Solow, R. (1974). "The Economics of Resources or The Resources of Economics." American Economic Review. Vol. 54, No.2: 1-14.

Stiglitz, J. (1976). "Monopoly and The Rate of Extraction of Exhaustible Resources." American Economic Review. Vol. 66, No.4: 655-661.

Taylor, L. (1979). Macro Models for Developing Countries. New York: McGraw-Hill.

Taylor, L. (1981). "South-North Trade and Southern Growth. Bleak Prospects from the Structuralist Point of View." Journal of International Economics. Vol. 11, No.4: 589-602.

United Nations (1982). Trade and Development Report, 1982. New York:

United Nations.

World Bank (1982). World Development Report. New York: Oxford University Press.

Cite this document

APA

Federal Reserve (1984, March 31). Oil Prices, Welfare Tradeoffs, and International Policy Coordination in a Three Region World Model: An Optimizing Approach. Ifdp, Federal Reserve. https://whenthefedspeaks.com/doc/ifdp_1984-241

BibTeX

@misc{wtfs_ifdp_1984_241,
  author = {Federal Reserve},
  title = {Oil Prices, Welfare Tradeoffs, and International Policy Coordination in a Three Region World Model: An Optimizing Approach},
  year = {1984},
  month = {Mar},
  howpublished = {Ifdp, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/ifdp_1984-241},
  note = {Retrieved via When the Fed Speaks corpus}
}