Expected Fiscal Policy and the Recession of 1982

International Finance Discussion Papers Number 272

December 1985

EXPECTED FISCAL POLICY AND THE RECESSION OF 1982

by

William H. Branson Arminio Fraga Robert A. Johnson

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to International Finance Discussion Papers (other than an acknowledgment by a writer that he has had access to unpublished material) should be cleared with the author or authors.

EXPECTED FISCAL POLICY AND THE RECESSION OF 1982

William H. Branson Professor of Economics, Princeton University

Armino Fraga Graduate Student, Princeton University

Robert A. Johnson Economist, Board of Governors of the Federal Reserve System

December 1985

We would like to thank Ivy Ellenberg and Cathy Crosby-Fitzgerald for research assistance. 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.

I. Introduction

The Economic Recovery Tax Act of 1981 had one aspect that is unusually useful for economic analysis. It provided an example of a clear-cut announcement of future policy actions at specified dates.

This provides an opportunity to apply recent advances in the analysis of expectations dynamics to cata that have been generated in an environ= ment. that includes such anticipated policy action.

| A three-stage future tax cut was announced in the Tax Bill in March 1981. In a Keynesian model with liquidity-constrained consumers or investors, or with uncertainty, this would normally be expected to provide a stimulus to the economy when the tax cuts actually appear. But the financial markets could look ahead to the stimulus and the shift in the high-employment deficit brought about by the tax cuts, and their implications for bond prices and interest rates. ‘In this paper we argue that this happened during the first half of 1981. As market participaitts came to understand that the tax and budget actions of March 1981 implied a future shift of the high-employment—now "structural"—deficit by some 5 percent of GNP, they revised their expectations of future real interest rates upward. This caused a jump in real long-term rates then, in 1981. And, we argue in section IV below, it also caused a sudden and unanticipated real appreciation of the dollar at the same time. The jump in rea]. long-term interest rates and the dollar appreciation in the first half of 1981 were essential features of the recession that began in July 1981.

This paper points out the possibility of a purely anticipatory

recession. If the only policy action had been the fiscal announcement,

and if ods markets are "Keynesian" but financial markets are forwardlooking, ‘the announcement can cause a recession, which will end when the actual fiscal action begins to stimulate the economy. In the actual context of 1981, a shift toward monetary tightness also contributed to the recession.

_ The models we use to analyze these policy changes have many antecedents. Wilson (1979) analyzed the effect of anticipated policies in Dornbusch's (1976) model of expectations and exchange rate dynamics. Buiter and Miller (1984 and references therein) have used similar models to analyze the theoretical effects of disinflation policies, as well as the actual events of the Thatcher period (1981, 1984). Miller (1980) and Blanchard (1981) have constructed closed-economy models which include an expectational term-structure of interest rates, and are simple enough, to be treated analytically. In particular, the initial model of ‘section III begins with Marcus Miller's (1980) four-equation closed-economy medel, and we have benefitted greatly from discussions with him on this topic (and many others). The paper is structured as follows.

In section II we begin with some "stylized facts" about the recession. The important things to notice are the sharp rise in teal longterm interest rates and the real appreciation of the dollar in early 1981, and the subsequent split of financing of the budget deficit between domestic saving and foreign borrowing--the current account balance.

Section III begins by incorporating a term structure based on expectations of future movements in interest rates into a standard fixed-price closed-economy IS-LM model. This is the model of Miller (1980). In it :

the short-term rate moves along the LM curve, but the long-term rate

follows a positively-sloped saddle path. This framework shows clearly the principles involved in analyzing expectations dynamics. “We then proceed to add a model of core inflation and a short-run Phillips curv: in order to incorporate price dynamics.

The open-econony version of the model is presented in section IV, with equations for the current account and an “open interest parity" condition with exchange-rate expectations and a risk premium that depends on the stock of government debt. This reflects our assumptiors that dollar-dominated and foreign-exchange assets are imperfect substitutes so that U.S. interest rates can move relative to foreign Yates. The open-econdmy model shows how the anticipated fiscal package could cause a jump in the real long-term rate and the dollar, splitting the financing between domestic investment and the current account.

The simulation version of the model is presented in section Vv. There we illustrate the dynamic’ effects of a variety of monetary and fiscal policies, and then combine them to produce a simulated view of the U.S. economy from 1980 to 1983. The results of this final simlation experiment are then compared with the stylized facts presented at the outset. Finally, we conclude with a discussion of the strengths and

shortcomings of our approach, highlighting a number of policy events “ whdeh occurred during this period, but are not integrated into this

exercise.

II. Economic Policies and the Unusual Characteristics of the 1982 Recession OL,

The 1982 U.S. recession coincided with major shifts in monetary and fiscal policies. In this section we will present a brief description of the main policy changes, as well as of the corresponding movements in some key endogenous variables. In. particular, we will focus on the financing of the deficit, and on the behavior of prices, production, interest rates, and the exchange rate. . | . |

| In March of 1981 the new budget and tax package was announced. It included a major increase in defense spending to be phased in over several years, a three-stage tax cut to begin in 1982, and some cuts in non-defense spending. The package as a whole implied a growing "structural" deficit in the Federal Government's budget from $40 billion in the first half of 1981 to about $180 billion by the end of 1984. The financing of the overall government deficit can be seen in Table 1. As required by the national income identity, a budget deficit must be financed by a combination of excess domestic saving over domestic investment, and a current account deficit (net foreign saving). Until the third quarter of 1982 most of the deficit was financed by net domestic saving. Investment fell from a peak of $495.8 billion in 1981:3 to $377.4 billion in the last quarter of 1982. In 1983, however, as the recovery gained momentum and the dollar continued to appreciate, a larger share of the deficit began to be financed by foreign saving.

Unlike fiscal policy, it is harder to characterize monetary policy in this period of financial innovation and deregulation. Even though the ' monetary authorities made explicit their intention of reducing inflation,

the main monetary indicators did not exhibit signs of larger policy shifts.

Table 1: U.S. NATIONAL INCOME ACCOUNTS, 1979.1 - 1984.2 eee ih 904

ees

Net Net Government Foreign Domestic Private Private Deficit Saving Saving Investment Saving —_—_— eeeeeSeSeeSeSeSeeeSSSFSFSeeSeSS 1979.1 -22.1 -2.3 -15.4 415.1 399.7 2 ~20.1 5.4 ~17.4 428.3 410.9 3 -12.9 -1.6 -14.6 431.9 417.3 4 ~2.1 5.7 -15.6 416.3 401.2 1980.1 7.5 4.1 77.3 422.0 414.7 2 38.1 _ 76.7 — 43.0 294.3 437.3 -3 43.3 ~20.3 61.3 379.5 440.8 4 33.9 ~2.3: 37.1 411.7 448.3 1981.1 8.1 ~11.3 13.3 455.5 468.8 2 i 10.6 -0.5 13.1 472.1 485.2 3 | 25.2 0.3 37.8 © 495.8 531.6 4 63.7 -5.5 76.6 476.2 552.8 1982.1 79.7 4.8 91.2 422.9 514.1 2 . 81.2 -8.7 . 88.2 432.5 520.7. "3 127.0 24.8 - 99.6 * 425.3 524.9 4 175.3 21.9 149.2 377.4 526.6 1983.1 142.9 6.7 137.4 404.1. 541.5 2 114.4 33.0 84.9 450.1 535.0 3 133.5 41.5 96.7 491.9 588.6 4 129.3 59.1 75.0 540.0 615.0 1984.1 107.4 77.7 27.5 623.8 651.3 2 n.a. 83.5 N.a. 631.5. nea.

Source: CITIBASE

Net Domestic Saving = Gross Private Saving (GPS) minus ‘ Gross Private Domestic Investment (GPI)

The growth rate of the money supply (Ml as currently defined) decreased from an annual rate of 8.32 between 1976.3 and 1979.3 to 6.1% between 1979.4 and 1982.3. In particular, from the second quarter of 1981 to the third quarter of 1982 Ml grew at an average annual rate of 5.4%. Finally, if we correct the money supply for the introduction of NOW accounts on December 31, 1980, we find that for the year 1981 the adjusted figure drops from 6.4% to 2.32%, indicating a severe monetary squeeze.

The behavior of the main macroeconomic variables in this period is summarized in Table 2. By mid-1981 nominal short-term interest rates had risen by about 5 percentage points over mid-1980, and long-term rates by 3.5 points, even as the inflation rate began to decline. In the third quarter of 1981 the recession began, and short rates dropped slightly. Long rates, however, remained high, a fact that we will discuss at length below. Almost simultaneously with the announcement of future fiscal policy and the rise in interest. rates, the dollar started its appreciation, which has continued to the end of 1984. The recession began in the third quarter of 1981 and ended in the fourth quarter of 1982, with a decline in industrial production of about 11.52%.

- If tight money were the sole cause of the recession, we would have expected to see the short-term interest rate remain above the long-term rate. However, since the fourth quarter of 1981, the short rate has teen below the long rate. The severity of the recession, the inversion of

the term structure of interest rates, and the appreciation of the dollar in recession are the puzzle that we attempt to solve in the next two

sections, focusing on expectations dynamics and future fiscal policy.

Table 2: ECONOMIC INDICATORS, 1979.1 - 1984.2

1979.1 8.7 9.4 9.0 0.3 101.9 100.5 99.0 — 2 8.7 9.4 9.1 0.4 99.6 100.0 98.8

3 9.1 9.6 9.0 ~0.1 101.9 100.0 100.0

4 9.5 11.8 10.2 0.7 100.1 99.8 100.1 1980.1 10.0 13.5 11.8 1.8 100.0 99.6 100.6 10.2 10.0 10.6 0.4 100.0 91.6 98.3

3 10.3 9.2 10.9 0.6 102.9 94.5 98.4

4 10.2 13.7 — 12.1 2.0 98.2 98.4 99.4 1981.1 9.5 14.4 12.7 3.2 92.6 99.6 101.5 2 8.9 14.5 13.5 4.6 85.8 100.1 101.7

3 844 15.1 14.5 6.1 81.8 99.2 102.6

4 748 12.0 14.1 6.3 85.9 93.9 101.3 1982.1 7.0 12.9 14.3 7.3 83.7 92.7 99.9 2 6.1 12.4 13.7 7.6 82.0 90.8 100.1

3 5.7 9.7 12.9 7.2 78.6 89.9 99.9

4 4.9 7.9 - [0.7 5.8 78.0 88.5 99.6 1983.1 4.0 8.1 10.9 6.9 80.4 91.6 . 100.2 2 4.1 8.4 10.8 6.7 79.1 | 95.8 -102.5

3 3.5 9.2 11.8 8.3 76.2 100.7 104.4

4 3.0 8.8 11.7 8.7 75.7 — 102.4 105.7 1984.1 3.5 9.2 11.9 8.4 75.4 105.2 108 .3 2 3.0 9.7 13.2 10.2 107.3 110.2

Sources: GNP: CITIBASE Others: IFS Data tape, IMF.

INFL = % increase in PCE deflator over the same quarter a year earlier.

SR = T-bill 3-month. |

LR = T-bond 20-year.

RLR = real long rate (LR - INFL).

RER = index of real effective exchange rate using WPIs (down = $ appreciation) ; this is the inverse of the IMF index.

IP = industrial production, indexed to 1979.3 = 100.

Gi = GNP at 1972 prices, indexed to 1979.3 = 100.

IIl. The Basic Model, Closed Economy

We begin the exposition with the simplest fixed-price IS-LM model of a closed economy. The basic idea of how expectations dynamics in

financial markets can generate a recession from an expected fiscal expan-

sion can be easily outlined in this framework. Then we will go on to price

dyaamics and open~-economy aspects of the model.

IS-LM with Short— and Long-Term Interest Rates

The basic model can be stated in four equations

(1) d= ay + i+uf - 6(R -h): IS Curve,

| (2) r= (ty - m)/e: LM Curve, (3) y = W(d - y): Gradual Output Adjustment, (4) ER=R-¢c: _ Path of Long Rate, or. Term Structure.

Variable definitions are given in Table 3. Equation (1) gives aggregate demand d as a function output y, autonomous investment i, the exogenous component of fiscal policy £, and the real long-term (actually consol) bond rate. Consumption is assumed to be a function of current income, in Keynesian fashion. This is clearly an important assumption for our analysis of fiscal policy. If infinitely-lived consumers take into account fully all future tax liabilities, including those related to debt service, then a shift from tax-financing to debt-financing of government spending will have no effect on aggregate demand. See Barro (1974) for a discussion of this case. For a variety of reasons such as “ liquidity constraints and uncertainty regarding remaining years of life

(see Blanchard, 1984), we think the neutrality assumption is too strong.

The expected, or "core" inflation rate h is given exogenously in equation (1), and set at zero for the time being. It will be endogenized later. Equation (2) is the LM curve normalized on the short-term interest. rate r. The short rate is assumed to clear the money market at all times. In the LM curve m is real balances M/P, and € is the semi-elasticity of the demand for money with respect to r. Equation (3) gives the change in output over time as a partial adjustment to the excess of demand over output. A more precise model would include inventory dynamics, but the specification here is sufficient to maintain a focus on expectations:

Equation (4) specifies the term structure of interest rates, providing one link with the future, and thus bringing expectations dynamics into the nodel . Aside from a risk premium, which we set to zero, any long-short differential must be equal to the expected rate of change of the long rate. If R - r>0 in equation (1), then the long rate mst be expected to rise (i.e. consol price to fall) to generate a capital loss that offsets the rate differential, for the bond market to be in equilibrium.

The dynamics of the model are described in the IS-LM diagram of Figure 1. The short-term rate r and the long-term rate R are measured on the vertical axis; output is on the horizontal axis. The stationary equilibrium is at point E, where R= r and d = y. Away from equilibrium, y and R move along the "saddle path" RR, and r moves along the LM curve.

The innovation in Figure 1 is the RR saddle path. This comes from the combination of equation (4) and the assumption of rational expectations in financial markets. In this non—stochastic model, rational

expectations is the same as perfect foresight. Output adjusts toward the

10

IS curve, following equation (3). This gives the horizontal arrows in Figure 1. The short-term rate clears the money market, so it moves along the LM curve following output. .

The RR saddle path can be derived as follows. I£ the long rate R were to adjust along the LM curve, ER would be zero throughout, from equation (4) with R= xr. This is incoasistent with rational expectations; to the left of E, r and R are expected to rise and to the right of E to fall. Therefore, to the left of E, for example, with ER > O, R mist exceed r. This gives the vertical arrows that show the motion of R. If R is .above LM, R > r and ER mist be positive for equilibrium between short and long rates; vice versa below LM. There is then only one saddle path RR that is consistent with a rational expectations equilibrium. It is positively shaped but flatter than LM. Along it, to the left of point E, ER =R-r>0 and to the right of E, it is negative. Other paths.of R,y are "bubble" paths © that cross IS vertically (y = 0) or LM horizontally (ER = 0) and explode to the northwest or southeast. ‘These “unstable branches" will play an important

role in the analysis of anticipated fiscal policy.

Unanticipated Monetary and Fiscal Policy

As a prelude to an analysis of the 1982 recession, we can use the diagram of Figure 1 to characterize the effects of unanticipated monetary or fiscal policy on demand and the term structure of interest rates.

This is the usual textbook case. Consider first the effects of a contractionary monetary policy (dM 0), illustrated in Figure 2. The LM curve shifts up, so the equilibrium moves from point A to point B. At the

initial levei of income Y? the two interest rates rise to Th Ry

-9Ja-

Figure 1: Basic Expectations Dynamics

-10a-

———— —. .

Yo

Figure 2: Monecary Concraction

-10b-

Figure 3:

=xpansionary Fiscal Policy

11

with the short rate above the long. As y falls, the two interest rates also fall, converging to point B.

The effect of an unanticipated fiscal expansion is shown in Figure 3. Again the equilibrium point moves from A to B. With y at va initially the short rate remains at rT. But in anticipation of the future rise in both rates, the long rate jumps to R. As output then rises to yy? beth

interest rates rise to Ry T°

Expected Fiscal Folicy

The Reagan Administration's future budget package, which implied a rising structural deficit, was announced in March 1981. The potential recessionary effects of an announcement of a future fiscal expansion are illustrated in Figure 4. With the economy at point A, a future fiscal expansion is announced. The financial markets come to understand that the future equilibrium is at point C, with higher interest rates. _. This means the long rate will jump immediately. But onto what path? The future saddle path will be RR at the time of the actual fiscal expansion. The long rate will rise seeking an unstable branch relative to the existing equilibrium A that has the following property: as the economy moves along that unstable branch, it will reach the new saddle path at the time the announced fiscal expansion actually takes place, i.e., when equi- Librium C comes into existence.

Thus the long rate jumps to Ry in Figure 4 with output at Ya: This depresses investment and sends the economy into recession along the

unstable branch from point 1 to point B. Output and the short rate fall:

to vy» Th, wnile the long rate rises to Ro: When the actual fiscal

12

expansion occurs, the recovery begins. Output increases from My, to Yo» with the short and long rates rising to converge to C. The financial]. market's anticipation of the future fiscal expansion raises the ptesent long-term interest rate and throws the economy into an anticipatory recession.

Anticipation of the future effects of. the 1981 budget and tax package cannot be the only cause of the 1982 recessicn, however. In Figure 4, the long rate rises and the short rate falls in the recession, and the short rate is below the long rate throughout. However, in the data of Table 2 we see that the short rate rose above the long rate in late 1979 and again “ror the fourth quarter of 1980 to the third quarter of 1981,

after which it has remained below the long rate. So current tight

monetary policy in 1980-81 must also be part of the story.

Tight Money and Anticipated Fiscal Ease 7 -

To explain the movements in the term structure in 1980-84, as well as the recession and recovery, we need a scenario that combines a monetary policy tightening in 1981 with the effects of anticipated fiscal ease.

In fact, growth of Ml slowed to 5 percent, annual rate, from December 1980 to July 1982, so this may be a fairly accurate scenario.

The combination of an actual tightening of monetary policy and an anticipated easing of fiscal policy is illustrated in Figure 5. For expositional convenience, we assume there that the two policies have an offserting effect on demand in the final equilibrium, that is, that after the recession and recovery, real GNP is back to its initial level rela~‘

tive to trend.

-l2a-

Expected Fiscal Expansion

Figure 4;

In Figure 5, the shift of the LM curve is current and unanticipated, while the IS shift is announced for the future. This creates a new current equilibrium at point B, and a future equilibrium at point D with a saddle path RR into it. At the initial level’ of income Ya? the short rate jumps to Ty on the new LM curve. The long rate rises, seeking the unstable branch relative to the new equilibrium B that will take the economy to RR when the IS curve actually shifts. So the long rate rises to point 1 to put the economy on the recessionary »ath From 1 to C.

Now we see that initially the short rate moves above the long rate. In the recession, as y falls from ys to Ye the short rate falls along the LM curve to The moving below the long rate at the bottom of the latter's path from l to C. The economy bottoms at point C when the actual fiscal stimulus comes on line. During the recovery period the two rates rise, converging to point D.

The essential feature of this scenario is the reversal of the term structure. At first short rates rise above long rates, and then fall below them, bottoming when the recession bottoms. Then in the recovery both rise with the long rate above the short. This is essentially the pattern of the data of Table 1. In the fourth quarter of 1980 the short rate jumped above the long rate. The crossover came in the first quarter of 1982, and the short rate remained below the long rate after that, bottoming with the recession in the fourth quarter of 1982. Both rates rose after that. Thus the scenario of an actual monetary tightening combined with anticipated fiscal ease is consistent with the broad move--

ments of the term structure and GNP.

14

Price Dynamics ‘,

The next step is to add price dynamics to the model. We adopt a medel of "core inflation," in which inflation adjusts gradually to.. monetary disturbances and is also sensitive to output disturbances. We use this model to reflect the idea that inflationary expectations are adaptive, rather than forward-looking. In our specification, the inflation rate is a geometric average of past money growth rates. This can be taken to represent a credibility effect, where a policy change is not immediately assumed to be permanent, as well as an element of stickiness on the supply side, such as would be implied by staggered wage contracts. People believed in the early 1980s that inflation was coming down only

|

\ when they saw it come down.

The inflation equations are two:

(5) -h=q (M -h): Adjustment of Core Inflation,

(6) p=h+d(y-y): Phillips Curve.

Equation (5) has the core inflation rate h adapting to deviations of money growth from h. Equation (6) says the actual inflation rate is the core rate plus a Phillips-curve term for deviations of output from its natural level y- For solutions of the dynamic model we will turn to computer simlation after introducing open-economy aspects in section III. But the solution algorithm for the dynamic closed-economy model is clear. The ISLM equilibrium is on a trend inflation rate given by M. Core inflation is M so h in (5) is zero. Equilibrium output is y, so p in (6) is equal? to h.

Demand and output are equal so y in (3) is also zero. With 2 = r from

-13a-

Rie . __- “N x ‘“ - , ‘ > Raaty > Ss ‘ ' ‘ | N

' et p o /

_——~ ne ee

Y 7 X

os a ° ‘ -|———~—— 4 me eee cece.

Figure 5: Actual Monetary and

Expected Fiscal Policy

15

(4), we can solve (1) for R and then (2) for P: ‘The saddle path into this solution gives the motion of the real long-term interest rate R=-h. The jump in the long-term real interest rate over the first half of 1981, and its continued upward movement since the bottom of the recession in the fourth quarter of 1982 can be seen in the data in Table l. The is consis-— tent with an interpretation of Figure 5 as showing the path of the real

long-term interest rate.

IV. The Oven Economy

{In addition to the historically high level of long-term real interest rates since the first half of 1981, the economy has experienced an appreciation of the dollar, in real effective terms, of some 30 percent since then. The data, using the IMF's index, are shown in the fifth colum of Table 2. Since we define the exchange rate as U.S. dollars per unit of foreign exchange, an appreciation means the real effective rate in Table 2 goes down. Again, the major movement in the exchange rate comes across the first half of 1981, consistent with the movement of real interest rates. To build this into the model, we have to open it up.

Trade Flows

We assume gradual adjustment of real net exports x toward an equilibrium level x that is a function of competitiveness, domestic income, and foreign income. Competitiveness is measured by the exchange rate

relative to the domestic price level (e/p), with the foreign price level.

tixed at unity. The equations for trade flows are:

16

me

(7) (8)

=ce/p + Ay* = wy : Equilibrium Trade Balance,

Woe

=y(x-x): Gradual Trade Adjustment.

The partial adjustment model is used to reflect the assumption that aujustment of trade flows to changes in competitiveness takes time. Eona-

ticn (1) should be rewritten to include the trade balance:

C1") ds ay+i+t pf + 6(R -h) +x.

Asset Markets

The central aspect of our model of exchange-rate determination is imperfect substitutibility between assets denominated in U.S. dollars ind in foreign exchange. With imperfect substitutability, an accumulation of U.S.-government debt can increase U.S. interest rates relative to "“wox id" interest rates by increasing the risk premium on U.$. bonds. With rational expectations, financial markets can look ahead to this implication of a shift in the full-employment deficit, and move the interest rate and exchange rate at the time when the implication becomes clear.

Imperfect substitutability and a risk premium that is positively

related to the bond stock are given by (9) Ee = r — (r* + pb),

where @ is the rate of change of the exchange rate, r and r* are the i} .S. and world short-term interest rates, and b is the real bond stock. This is the standard open interest parity condition with a flexible exchange:

rate and a risk premium.

17

The real. bond stock accumulates as the nonu-monetized part of the

budget deficit, following (10) b= £ - MM

where f is the current real deficit, wis the growth rate of money,

and L is real balances. © ' Agents in financial markets are assumed to look ahead to the consequences of shift ts in f or b through equation (lo). . |

With perfect foresight, equation (9) Says that the current level of the exchange rate must reflect the integral of future expected interest differentials, adjusted for the risk premium. In the solution of the open-economy nodel, agents look ahead to the path of r — r* - pb. Throughout the analysis, we hold r* constant. If this integral is positive, the expected value of e@ must be positive, so the current exchange rate must fall below its equilibrium value given the current fundamentals. Thus for sufficiently small p, the announcement of a future fiscal deficit combined with a current shift to tighter monetary policy, by increasing the expected future short-term interest rate, will yield an appreciation of the dollar, at the time of the announcement. Buiter and Miller (1983)

give the forward integral of (9) as

ao

(ll) e(t) = e(t) + | E.(r* + pb(s) ~ r(s))ds, t

where e(t) is the current long-run equilibrium exchange rate. When the expected integral becomes negative, e(t) falls relative to e(t) so the

expected rate of increase of the exchange rate,e(t), is positive.

Long-run Equilibrium

The long-run equilibrium of the open-economy model can be solved as follows. From equation (5) with h = 0, the core inflation raté-is h= M, which is fixed exogenously. From (6) with y= y> pP=h also, and m = M/p is constant. From (3) with y = 0, d= y. In long-run equilibrium the real exchange rate e/p will be constant, so e = p =h = M. Combined with the long-run bond stock b, this gives us the equilibrium value of the short rate r from (9), and from (4) with R = 0, the long rate and the short rate are equal. With M given exogenously, and y and r determined, the price level P comes from money-market equilibrium (2), and the 1s equation (1) can be solved for net exports x. With x = 0 in (8), X = x, and finally equation (7) can be solved for the level of the nominal exchange rate e that yields a real exchange rate (e/p) consistent with equilibrium x. This completes the long-run equilibrium

solution.

Intuitive Dynamics of a Temporary Fiscal Expansion

The dynamics of fiscal and monetary policy will be studied using a simulation version of the model in section V. Here we will outline the intuition behind the results using the example of an unanticipated and temporary fiscal expansion. Te must be temporary so that the real bond stock reaches a new long-run equilibrium value rather than going off to infinity. So consider a temporary increase in f that creates a permanent increase in b, with no change in money growth. This fixes long-run

a

nominal growth rates e = 5 =h = M.

Figure 6: Short-run Adjustment with

Temporary Fiscal Expansion

J8

19

First we consider long-run equilibrium. With a larger bond stock and no change in the equilibrium e, rand R mist rise, from (9) and (4). The rise in long-run equilibrium r means the entire price path must. rise to reduce real balances. With a higher real interest rate R - h, net exports must rise, so e/p must rise. This implies that the e path rises more than the p path. So in the long run interest rates rise and the exchange rate depreciates.

Now let us consider the shorter run, while the budget is in deficit, but y= d= y- In this short-run equilibrium, some combination of an increase in the real interest rate (R ~ h) and decrease in (e/p) are required in order to reduce investment and net exports to make room for

the increase in f. Total differentiation of equation (1) with y and d

given, p set at unity, and (7) for x results in

(12) udf - 6dR+ cde = O.

In the R,e space of Figure 6, equation (12) is the LX curve with slope o/é and an upward shift when f increases. But in the short-run equilibrium r = R, and equation (11) says that if market participants expect r to rise, the exchange rate will immediately fall. This gives the negatively-sloped am (for asset markets) line in Figure 6. The two conditions together give the increase in R and decrease in e that split the financing of the f deficit between investment and net exports in the short run. This is a partial explanation of the increase in real interest rates and appreciation of the dollar that was shown in Table 2, in order to obtain the shift in the private saving-investment balance and net :

exports that was shown in Table l.

20

V. Dynamic Policy Simulations

In this section we will examine the results of a variety of dynamic simulation experiments using the model developed in the previous sections. For convenience, this model and the parameter values we have used are reproduced in Table 4. The simulations will be performed using the computer program of Austin and Buiter (1983), as modified by Johnson (1985). The program simulates linear, dynamic, perfect-foresight models applying the methods developed by Dixit (1980) and Buiter (1984).

First we will compare and contrast the impact of anticipated versus unanticipated fiscal stimulus. We will then explore the implications of monetary conttaction, alternatively modelled as an unanticipated fall in the rate of growth of money and as an unanticipated fall in real balances. Finally, we will combine the elements of unanticipated monetary contraction followed by anticipated fiscal stimlus and compare the dynamic paths of the variables in our model with the data series which were dis-

cussed in section ITI above.

Unanticipated Fiscal Stimlus

Figures 1.1 -1.3 illustrate the result of a fiscal expansion which is unanticipated. The stimulus to aggregate demand and the concomitant increase in the future supply of bonds to finance the deficit lead to an immediate jump in the long rate of interest as the policies are announced. The short-term rate of interest rises as the increase in demand increases

the transaction demand for money. Figure 1.2 shows that the nominal

21

Table 3: VARIABLE DEFINITIONS ‘

d aggregate demand -y output i autonomous investment

£ budget deficit

R long-term interest rate

x short-term interest rate

h expected or "core" inflation

x trade balance

x | equilibrium trade balance

m ' real balances

M money supply

Pp price level

e exchange rate . b stock of real bonds

E conditional expectations operator

Note: a bar over a variable denotes the steady-state value, a hat denotes proportional rate of change, a dot the time derivative, and a star a

foreign (and exogenous) variable.

22

‘\ Table 4: THE COMPLETE MODEL \

(1) d = ay+it+yf - 6(R-h) +x | . Parameter values

(ty - a)/e +r4

(2) r* a = .80 $ = .80 (3) y = Wd -y) us 1 a Y m= 1.00 (4) ER *= R-o¢ y= -80 , g = .15 (5) h = 7(M - h) (6) Pp = h+o(y - y) 0.2 -! / _ R = 110 (7) x = oe/p + ly* - wy ty e E€ = (8) x = (xX - x) p - 0 A ¢ = .50 (9) Ee = rx - (r* + pb) . Tr = 8.0 (10) bo = £- SL U-d -Mm te r zx .02

Note: For simulation purposes, equation (10) in the text was linearized to

obtain equation (10) here.

23.

exchange rate and, because of sticky prices, the real exchange rate’

jump appreciates upon announcement of the fiscal stimulus and then depreciates continuously. The trade balance deteriorates following the’ real exchange appreciation and gradually offsets the impact of fiscal stimulus upon aggregate demand. Output y is shown in figure 1.3. Initially output rises as the direct fiscal effect dominates reduced demand in the interest sensitive and foreign sectors. Thereafter the latter two effects

predominate and the economy slackens.

Anticipated Fiscal Stimulus

Figures 2.1 - 2.3 exhibit the results of the simulation of a fiscal expansion announced one year before its actual implementation. Figure ‘2.1 shows that, just as wnen the fiscal stimulus is unanticipated, the long rate of interest jumps in anticipation of future bond supply increases. A comparison of figures 1.1 and 2.1 reveals that when -economic agents anticipate the fiscal expansion one year before implementation, the resulting jump in the long-rate is smaller upon announcement. Similarly the appreciation of the real exchange rate is smaller initially when the announcement precedes the implementation of the policy. This can be seen by comparing figures 1.2band 2.2b,

The impact of the announcement of future fiscal stimulus upon current output is snown in figure 2.3. Unlike the immediate fiscal expansion shown in figure 1.3, anticipated expansion produces an “expectations recession" with output initially falling in response to both the rise in interest rates and the deterioration of international competitiveness. :

It is only after the implementation of the expansionary fiscal policies

24

that output begins to recover. This trajectory of output in response to expected future fiscal expansion is attributable to the contrasting assumptions employed in modelling the behavior of financial markets with actions in the goods and factor markets.

We have assumed in our specification that agents in the real economy "believe what they see" whereas in the financial markets, participants “see what they believe." Thus an anticipated fiscal expansion may produce movements along a demand function as the cost of capital increases but the expected policy does not produce a shift in demand prior to the imp léementation of the policies either because of increased expected profitability or, in the case of consumption, because of expected increases in future income. |

We will now turn to the examination of the impact of monetary contraction upon the variables in our model before attempting to compare our. fiscal experiments with the time series data discussed above. Following that, we will model the combined influence ‘of monetary and fiscal policies with a better understanding of how each policy instrument

contributes to our simulated view of the causes of the recession of 1982.

Reduction in the Growth Rate of Money i neha

The simulation of an unanticipated permanent reduction in the rate of growth of money from 5.3percent to 2 percent is plotted in figures 3.1 - 3.3. Figure 3.1 reveals that upon announcement the long rate R

jumps downward, reflecting the expectation of falling interest rates in’

the long run, as some of the inflation premium is "wrung out" of nominal

25.

interest rates. The short-term rate does not move discontinuously from

the initial position of 8 percent, reflecting the fact that a change in

the rate of money growth does not imply an instantaneous change in. che

stock of money and that prices and income adjust gradually. As the new

lower growth rate of money translates into a smaller nominal money stock” ~~ ~~~ while prices adjust sluggishly, the short rate is driven up as real

balances decline.

The magnitude of the rise in short rates is also increased because aggregate demand is stimulated by the fall in long interest rates and produces a greater transactions demand for money in the short run. .

Figure 3\2breveals that the real exchange-rate jump appreciates in response co the announcement of reduced money growth. This jump appreciation corresponds to a jump in the nominal exchange rate to a new “dynamic path because of the expected lower path of prices.

Figure 3.3 plots the path of output after the fall’ in the growth rate of money. Initially output is stimulated by the jump decline in the long rate of interest. Thereafter the downward adjustment of inflationary expectations increases the perceived long real rate of interest and at the same time the trade balance deteriorates following the initial exchangerate appreciation. In the long run the real rate of interest is restored to its equilibrium value and the trade balance improves following a

depreciation of the real exchange rate.

Fall in the Level of Real Balances M/P

A second characterization of a monetary contraction consists of an:

unanticipated fall in the level of the money stock without altering the

26

steady-state rate of money growth. Figures 4.1 - 4.3 exhibit the results of an experiment where the stock of money falls at time t = 0 by 3 percent. |

In Figure 4.1 we see that the inversion of the term structure, as measured by the difference between the long and the short rates of interest, is primarily attributable to a jump in the short rate though the long rate also moves at the time of the announcement.

On the other hand, the inversion of the term structure of interest rates in figure 3.1 was primarily attributable to a fall in the long’ rate of interest in anticipation of falling inflation. In this simulation steady-state jinflation does not decline.

Figure dt monce again reveals that the real exchange-rate jump -appreciates due to the expectation of a lower price level brought about by transitory disinflation resulting from the money’ stock decline. Thereafter the real exchange rate depreciates back toward its initial level.

The trade balance, shown in figure 4. follows the exchange rate with a lag. Figure 4.3 shows that, unlike a money~growth rate decline, the drop in the stock of real balances produces an unambiguous immediate decline in aggregate. demand as both the real exchange appreciation and

the rise in interest rates depress expenditure.

Policy Simulation and the Recession of 1982

In this section we combine the various monetary and fiscal policy influences in a manner which we feel roughly corresponds to the macroeconomic policies which influenced the U.S. economy between the third

quarter of 1979 and the end of 1984. The principal policy events which

27

we try to capture are:

1. A contraction of the money stock in October of 1979.

2. A tendency for the monetary authorities to reduce the growth rate

of money over the period.

3. An anticipated three-stage fiscal expansion announced in March of

198i to be implemented over the years 1982-85.

For the purpose of the simulation experiment we have sought to capture these policy influences as follows. We begin the simulation in October of 1979 by implementing an unanticipated drop in the money stock. Then, in March of 1981 we simultaneously introduce a3.3percentage-point reduction in the growth rate of money and a three-stage anticipated fiscal expansion to be implemented th steps, 3, 7, and 11 quarters in the future. The results are shown in figures 5.1 - 5.3 and can be compared with the data for the U.S. economy from Table 2 above.

Initially the short rate of interest increases by over 100 basis points while the long rate rises by 20 basis points. The. inversion of the term structure persists until the announcement of a fiscal stimulus produces a rise in the long rate in anticipation of rising interest rates due to bond finance of the prospective deficit. ;

Figure 5.2hshows the pattern of the real exchange rate and can be compared with the fifth column of Table 2. The marked event is the large real exchange-rate appreciation upon announcement of the fiscal program and monetary deceleration. In our stylized view this event roughly corresponds to the period between the election in the fourth quarter of 1980 and the third quarter of 1961 when the fiscal program of the Reagan Administration was formed and the information diffused into the market-

place.

28

Figure 5.2Ashows the deterioration of the trade balance over this period. Figure 5.3 reveals the impact of the simlated policies on output. The money squeeze in 1979 induces some recessionary behavior but the severe "expectations recession" begins with the announcement of the fiscal program in 1981 and reaches bottom at the end of 1982 with particu-

larly deleterious effects on the foreign and interest sensitive sectors.

29

In the final simulation experiment conducted above we have chosen a particular configuration and timing of changes in monetary and fiscal policy instruments to represent a stylized view of U.S. macroeconomic policy in the early 1980's. The implications of these actions for key simulated variables, given our structural specification, were then compared with the actual historical behavior of those variables, to assess the correspondence between the two series.

The quality of our "fit" in the simulation especiaent is affected by several limitations which we would like to stress briefly. They fall into two groups: methodological limitations, and omitted events.

The dynamic simulation methodology utilized here constrains one's modeling of forward-looking events by requiring that upon announcement of a new policy environment, stretching indefinitely into the future, agents have complete and immediate perfect foresight. Thus, new policies are completely credible upon announcement, and there is no time lag between the announcement and "belief," because of lags in the diffusion of information.

In our experiment this methodological restriction implies that changes in such variables as the term structure of interest rates occur in discrete and discontinuous jumps as new information is revealed. For instance, the announcement of the future fiscal expansion in 1981 produced an instantaneous jump in long rates of interest, whereas the historical data on Table 2 show

that the rise took place gradually over the second and third quarters of

1981.

While we are comfortable with the assumption that new information

propagates through the financial markets relatively nore quickly than through

30

goods and factors markets, we feel that the assumption of instantaneous perfect foresight imposed by the methodology introduces distortions into the simulation.

The other source of discordance between simulation and historical variables is the omission of several key aspects of the policy environment over the period under study. In particular, in order to retain simplicity and facilitate exposition we have omitted some important financial events. The Credit Control program of 1980 introduced limitations on the behavior of participants in U.S. financial markets which we have not incorporated into the simulation. Similarly, we have not introduced the deregulation of interest rates on instruments included in money aggregates, nor have we sought to eal with any aspect of what might be termed "financial innovation." Finally, in this version of the Paper we have not modeled the so-called "safe-haven effect" or dealt with the rapid growth in the money supply, both of which occurred in the third quarter of 1982. Studies of these additions to our stylized view of policy events constitute an agenda

for future research.

31

References

Austin, G. A. and Willem H. Buiter (1982), "'Saddlepoint': A Programme for Solving Continuous-Time Rational Expectations Models," University of Bristol Working Paper No. 82/132, November.

Barro, Robert J. (1974), “Are Government Bonds Net Wealth," Journal of Political Economy, 82:1095-1117, November/December.

Baumol, William J. (1959), Economic Dynamics, MacMillan.

Blanchard, Olivier J. (1984), "Debt, Deficits and Finite Horizons," NBER Working Paper No. 1389, June. -

(1981), “Output, the Stock Market, and Interest Rates," American Economic Review, 71:132-143, March.

Buiter, Willem H. (1984), "Saddlepoint Problems in Continuous-Time Rational Expectations Models: A General Method and Some Macroeconomic Examples," Econometrica, 52:665-680, May. ; cone oS

and Marcus H. Miller (1984), "Real Exchange Rate Overshooting and the Output Cost of Bringing Inflation Down: Some Further Results,"

in Jacob Frenkel (ed.), Exchange Rates and International Macroeconomics, University of Chicago Press, Chicago.

(1983), "Changing the Rules: Economic Consequences of the Thatcher Regime," Brookings Papers on Economic Activity, 2:305-365.

Dornbusch, Rudiger (1976), "Expectations and Exchange Rate Dynamics," Journal of Political Economy, 84:1161-1176, December.

Dixit, Avinash K. (1980), "A Solution Technique for Rational Expectations Models with Applications.to Exchange Rate and Interest Rate Determination," mimeo, University of Warwick, November.

Johnson, Robert A. (1985), "A Note on the Solution of Linear Saddlepaths when the Forcing Vectors are Modelled as a Step Function," mimeo, Federal Reserve Board.

Miller, Marcus H. (1980), "What Does Forward-Looking Behaviour in the Gilt's Market Imply for (a) the Price of Consols and (b) Monetary and Fiscal Policy?" mimeo, University of Warwick, December.

Wilson, Charles A. (1979), "Anticipated Shocks and Exchange Rate Dynamics," Journal of Political Economy, 87:639-647, June.

VL 3yNdIS ; C66l IGGL OGE6l GSE! SSEl LEGl 9BEl SBEL FREI SEL 2861 IREL O10°0 0z0°0 O£0'0 Ov0'O OSO'O F.._._ 18. n eee eee ae an - 0390'0 es a OL0°0 0g0'0 | J

O600F anne a oo

nae wow oe eS

lor ne) el'O lie) vl'O ue

NOILVIANI = --—--—--- _

30 JLVa ISSYSLNI LYOHS ZVO 3lvy¥ LSSYSLNI SNO?}

=] 8)

NOISNVdX3 WOSId GALVdIDILNYNN NOILVIANI GNV SS3lVa LSSYSLNI GALVINNIS

Vel Sends

c66l 1661 O66! 686l BSEl LE6l YSEl SBé6l VB6l S86 cl 186l

NOISNVdX3 IWOSIsa GALVdIDILNVNN SONVIVE IWNYSLXS GALV INNIS

S*0

9°0

L0

del 3YNIIS

C66! 1661 O66l 6861 BRél L86l G86l S8El PEEl SB6l ZBEl LEI

NOISNWdX3 WOSIS GALVdIOILNVNN JiVY SONVHOX]S We G3LvVINNIS

66°0

OLOL OcO'l OLO'l OvO'l

OSsOo'l 090"! OZL0'1 Oso’l 06011 Lh

bh

c66l 1661 O66! 686l BSEl LE6l YSEl

——_ oa ot

ft SYNdIs

S86l 786l S8é6l c8é6l IB6l

GNYW3G S3Lv9aYu9Dv = | INdiNo = ———___

NOISNVdX4 WWOSI4 G3ALVdIDILNWNN ONVW3Ad0 S1V93SY499V GNV LNdLNO GALV INNIS

O O10°0 0z0°0 O£0'0 Or0'0 0SO'0 090°0 OL0°0 080'0 060°0 Ne LO Ame) LO ane Slo

re 3YaNDIS

c66l 1661 O66l 686! SSEl LE6l YB6l SSEl VEG! SSeE6l CEL IBEl

NOLWWISNI = -—-—-—- - Jive LSSYSLNI LYOHS = ------- 31Vy¥ LSSAYSLNI SNOT = —————

NOISNYdX3 WWOSIS G3SLVdIOILNY NOILVIANI ONV SS3LlVY LSSYSALNI G3ALV INNIS

Vee SENSIS

2661 1661 O66! 686! SVEl LE6l GSE SBEl rE6I SSEl ZREl 186

NOISNVdX3 IWOSI4 G3LVdIDILNY SJONVIVE IWNYSLXS GSLVINNIS

LO

8°0

6°0

el

v'l

G‘l

gee 3YNSIS

Z66l l661 O66! 686! SS6l LBEl SBEl SBEl FEE SBEl 7BéEl 1BE6l

NOISNVdX3 IWOSIS GALVdIOILNY diva AONVHOXS Way GSaLV INNIS

ge 3YNDIS

60 c66l 1661 O66! 68S6l SSE LB6l SB6l SSE VB6l SSé6l CBE LBé6l 16°O c6'0 °6°O v6'0 I S60 io 96°0 4

L6°0 I \ 86'°0 66°0

OLO’L FAS 2b ~ 5

OcO0'l >> OL0't | ON ten HO Ovo'l

OSO'l

090’ GQNVWSO S31vV93SuDDV = ----~--- OLO0'l LINdino =

O8O0'l

— me me

NOISNVdX3 WOSI4 AG3LVdIDILNY ONVWAd SLV9RYNSD9V GNV LAdLNO G3ALVINNIS

re 3Y8NSI4 - C66! 6861 B8EI LBEl 9BEI SBE! HEE S86! ZB6I 1861 OBE! LE! OOO F ee wT

0Z0'0 F-----~---- ninee .

O£0'0 . Ov0'0 0S0°0 090°0 OL0°O 0g0°0 060°0 "0 "0 ZL'O S10 vl'0 Sl'O | NOILVTANI = -----~-- -

0 3lVy LSSYSALNI LYOHS LV'O J1lVva LSSYALNI SNOT

80

HLMOYD AFNOW 4O 3LVY SHL NI ANITIOSG G3LVdIDILNYNN NOILVIANI OQNV SSALVY LSAYSLNI GSLVINWIS

020°0 090°0

0SO°O-

Ov0'O O£0°O 0c0°0

O10°O

O10°0 0z0'0 O£0'0 Ov0'0 0SO'0 090°0 OL0°0

vee SYNSIA

O66! 68El BSI LEE! YBEl SSE PEI SSI ZEI 1861 OBEl 6LEl

HLMOYD ASNOW SO S1VY SHL NI SNITOR0 G3ALVdIDILNYNN JONVIVE WWNYSLXS GALV INNIS

gee 3YNSIS

0661 686! SR6l L86l SVEl SB6l PEG! SBE CBEl IGEL OBEL 66

ol Sol —_——— aa | Go"l vl Stl SG‘

HLMOYD AANOW 3O 3LVY SHL Ni SNITDSG G3LVdIDILNWNN diva AONVHOX]A Way GSLVINNIS

L6°0 c6'0 £6°0 - v6'0 S6’0 96°0 £6°0 86°0 66°0

OLO'l OcO'l OLO't OvOo'l OSO't oso OLO'1 Oso’

ge 3yNIIS

O66! 686! B86 L861 SSEl SB6l PVe6l S8é6l cBE!l 186! OSEl 646!

GNVN30 SLVSSYdDSV. = ----indino =————

HLMOYD ASNOW SO 31Vy4 AHL NI ANIIORG dalvdlolLNVNN. OQNVW3d0 S1V9SEYDDV GNV LAdLNO GALVINNIS

*y 3YgNIIS

ooo O66!) 686! SEI LEEl OBE SEBEL PEG S8El BEI IBEl ORE! 6/6

Sv0'0 pe eee eee eee OSO'O }-------------- hanna ire SSO°O O390°0 S90°0 OL0°O SLO0°0 ee O80'0 + , S80°0 -~ 060°0 S60°0

LO SOLO | WO NOLLVISNI = -----~- -

3ivY ISSYSLNI LNOHS = --—---—-— SILO | 31vVa¥ LS3Y3SLNI SNOT =

AM) AOSOLS ABNOW SHL NI SNITO3G G3LWdIOILNVNA | NOILVIANI GNV S3SLVY LSSYSLNI GALVINNIS

Vey SYUND

__ O66 68E6l BBEl LE6I YVEl SBEI wel SEL ZBGI Gel OBEL LEI OZ0°0 a a

4z0'0 } ¥Z0°0 1Z0°0 — 8100 _ S10"0 | cL0"O | | \ 600°0, --—_—____—_ \ 900°0

£00°O

MOOLS ASNOW SH1L NI ANITDAG G3LVdIDILNWNN JONVIVE IWNYSALXA GCALV INNIS

Z2e'l See'l S2'l See'l vz'l Sre'l Sz SSe'l 92" S921 lel lel ge'l see'l 621 S621 ol

Ge°y 3YNIIS

O66! 686l BREl L86l 9B6l SBEl PEGI SSé6l ZEEI ISEl OGEL 6/61

MOOLS AANOW SHL NI ANITDAG G3LvdIOILNVNN Jive SONVHOX]A W3Y G3ALVINNIS

SL6'0

86'0 —§86'0 66°0

S66°0

SOOo'l OLO'L SlLOL

OcO'l

£’y 3YNdIS

O66! 686l BBéEl LZE6l 9BEl SBE6l VES! S8é6l cBEl 18El OGEl 6ZEl

GQNVW30 SlV9SYS9SV = .~- ~~~ LAdLNO

MOOLS ABNOW SHL NI ANITOSG G3LVdIDILNVNN ONVW3d S1V9RSYD9V GNV LAdLNO GALVINNIS

WS 3AdNDI4

O66! 6SEl SBEl LE6l BVEl SBEl PRE! SBE ZBEL 1GEL OBEL GEL i a ns es es es a

C10°O

020°0 O£0'O Ovo'o ~0S0°0 090°0 O0ZL0°0 O80°0 O60°0 LO LPO elo en EVO -- La FLO |- ae NOUYTAND 20 ---e ee a Jaivt) LSSYSLNI LYOHG + ----~-— GVO f- ALWY LSSYILNIDNOT - -———-——. - Oro L- | | | ~

SATIOWILS WOoOS CALVGIOLLNY AG JAMOTIOS NMOD TART ASINGW NGiILV IAIN UNV SSLIVU LSSYaLNi GALWTaWS

VEG FANG

O66 6REl BEL /BEl YBEL SBE! PAE! SEL SGI IGE: OBEL GL5I LV'0 se nn On Oe Gees ee ee es 310 | SIO rl'0 S10 AMO) LO "0 060°0 080'0 OL0°O 090°0 0SO'0 Ov0'0 O£0'0 0z0'0

SMINNILS WOSI4 G3LvVdIDILNY AG GAMO1TIOS NMOG 13A57 ASNOW JONVIVG IWNYALXS GalV INNIS

es JANI

O66! 686i B86I LEEl OBE SRE! PEGI SEI ZBEL IGEL OBE GLE

SNINWNILS WOSsl-4 QALVdIDILNV AG JAMONNO- NMOQG 153A37 ASNOW | JIVU SJONVHOXS “IVEY GALvInnis

£’S AYNDIz!

O66l 686! SB6l LE6l GV6l SGB6l VEEl S86l cel ISElL OBEl 661

S8°0 L£8°0 68°0 L6°0. £6'O S6°0 L6°0 66°0

O10

OL0'1

—Nen wo

OQNVWS30 S1V9SY99V LAdDLNO

OSO'l

OZLO'1

SMINWILS WOSId GAaLvVdIOILNY AG GAMO1103 NMOG 13A357 ASZNOW ONVW30 SLV9OSYSD9V GNV LAdLNO GALV INNIS

IFDP

NUMBER

259

258

257

256

255

254

253

252

251

250

249

248

247

246

245

244

International Finance Discussion Papers ee cssion rapers

TITLES

Trade Policy for the Multiple Product Declining Industry

Long Memory Models of Interest Rates, the Term Structure, and Variance Bounds Tests

Currency Substitution and the New Divisia Monetary Aggregates: The U.S. Case

The International Transmission of Oil Price Effects and OPEC's Pricing Policy

U.S. Banks' Lending to Developing Countries: A Longer-Term View

Conditional Econometric Modelling: An Application to New House Prices in the United Kingdom

Low Pushing: Doctrine and Theory

1984 (partial listing) Postwar Financial Policies in Taiwan, China

Foreign Exchange Constraints and Growth Possibilities in LDCs

The Determination of Front-end Fees on Syndicated Eurocurrency Credits

Monetary Policy Games and the Role of Private Information

The Macroeconomic Implications of Labor Contracting with Asymmetric Information

Cooperative Policies among the North, the South, and OPEC: An Optimal Control Approach

International Repercussions of the U.S. Budget Deficit

Can Exchange Rate Predictability Be Achieved Without Monetary Convergence?- Evidence from the EMS

Domestic Saving, Current Accounts, and International Capital Mobility

AUTHOR(s)

Catherine Mann Gary S. Shea

Jaime Marquez Jaime Marquez

Henry S. Terrell Rod Mills

Neil R. Ericsson Davida F. Hendry

William Darity, Jr.

Robert F. Emery Jaime Marquez

Rodney H. Mills Henry S. Terrell Matthew B. Canzoneri Matthew B. Canzoneri Anne C. Sibert

Jaime Marquez Peter Pauly

Peter Hooper

Kenneth Rogoff

Gerard Caprio, Jr. David H. Howard

IFDP

NUMBER

272

271

270

269 268

267 266

265 264

263

262 261

260

International Finance Discussion Papers

TITLES 1985 Expected Fiscal Folicy and the Recession of 1982

Elections and Macroeconomic Policy Cycles

Assertion Without Empirical Basis: An Econometric Appraisal of Monetary Trends

in ... the United Kingdom by Milton Friedman

and Anna J. Schwartz

Canadian Financial Markets: Proposal for Reform

Was It Real? The Exchange Rate Interest Differential Relation, 1973-1984

The U.K. Sector of the Federal Reserve's Multicountry Model: The Effects of Monetary and Fiscal Policies

Optimal Currency Basket in a World of Generalized Floating: An Application to the Nordic Countries

Money Demand in Open Economies: Substitution Model for Venezuela

A Currency

Comparing Costs of Note Issuance Facilities and Eurocredits

Some Implications of the President's Tax Proposals for U.S. Banks with Claims on Developing Countries

Monetary Stabilization Policy in an Open Economy

Anticipatory Capital Flows and the Behaviour of the Dollar

Simulating Exchange Rate Shocks in the MPS and MCM Models: An Evaluation

The Government's

AUTHOR(s)

William H. Branson Arminio Fraga Robert A. Johnson

Kenneth Rogoff Anne Sibert

David F. Hendry Neil R. Ericsson

Garry J. Schinasi Richard Meese Kenneth Rogoff

Hali J. Edison

Hali J. Edison Erling Vardal

Jaime Marquez Rodney H. Mills

Allen B. Frankel

Marcus H. Miller Arnold Kling

Arnold Kling

Please address requests for copies to International Finance Discussion Papers, Division of International Finance, Stop 24, Board of Governors of the Federal Reserve Board, Washington, D.C. 20551.