Transcript Macro Conference IV

Chapter 8
Exchange-Rate Management:
Contractionary Devaluation
and Real-Exchange-Rate Rules
© Pierre-Richard Agénor and Peter J. Montiel
1


Contractionary Devaluation.
Real-Exchange-Rate Targeting.
2
Contractionary Devaluation





Consider a small open economy that operates under a
fixed-exchange-rate system.
“Dependent economy” framework in which traded and
nontraded goods are produced using
 homogeneous, intersectorally mobile labor;
 sector-specific capital;
 imported inputs.
Production costs may be affected by the need to finance
working capital.
There are variety of mechanisms to determine the
nominal wage.
Households hold money, capital, and foreign securities
and issue debt to each other.
4



Effects on Aggregate Demand.
 Consumption.
 Investment.
 Nominal Interest Rates.
Effects on Aggregate Supply.
 Effects on the Nominal Wage.
 Imported Inputs.
 Effects Through Costs of Working Capital.
Empirical Evidence.
 Before-After Approach.
 Control Group Approach.
 Econometric Models.
 Macro-Simulation Studies.
5
Effects on Aggregate Demand

Demand curve facing the traded goods sector is given
by the law of one price:
PT = EPT*,
PT: domestic-currency price of traded goods;
E: nominal exchange rate;
PT*: foreign-currency price of traded goods (unity).
6

Aggregate real demand for nontraded goods:
dN = cN + IN + gN,
cN: domestic consumption;
IN: investment;
gN: government demand.
7
Consumption

Consumption demand for nontraded goods:
cN = cN(z, y - tax, i - a, a; ).
(1)
z = PT/PN: real exchange rate (PN is domestic-currency
price of nontraded goods);
y - tax: real factor income received by households net of
real taxes;
a: real household financial wealth;
i - a: real interest rate (i is domestic nominal interest
rate and a expected inflation rate);
: shift parameter to capture distributional effects on
aggregate consumption.
8
Effects of devaluation:
Relative Price Effects:
 Nominal devaluation causes changes in relative prices
 This affects demand for domestically produced goods.
 Total (foreign and domestic) demand for domestically
produced traded goods is perfectly elastic and thus is
not affected by relative price changes.
 But total (domestic) demand for nontraded goods is
affected by changes in relative prices.
 Devaluation has a relative price effect on the demand
for domestically produced goods through its effect on
the demand for nontraded goods.
9

Real depreciation of the domestic currency (increase in
relative price of traded to nontraded goods) increases
demand for nontraded goods.
10
Real Income Effects:
 Devaluations produce changes in real income that affect
the demand for domestically produced goods.
 Real-income changes can be decomposed into those
resulting from changes in
 relative prices at the initial level of output;
 output at the new relative prices.
 Price level:
P = EPN1 - , 0 <  < 1,
(2)
: share of traded goods in consumption.
11

Real income:
y = yNz - + yTz1 - ,


yN: production of nontraded goods;
yT: production of traded goods.
Effect of a real devaluation on real income for a given
level of output is ambiguous.
Differentiating (3) with respect to z:
dy/dz = z-1(-)(yNz -+yTz1 - ).

(3)
(4)
 is share of traded goods in total output:
 = zyT/(yN+zyT).
12





(4): impact effect on real income depends on whether
traded goods have a higher share in consumption or in
income.
Assume:
 no expenditure on investment goods;
 no public sector expenditure, so that cN = yN.
Then net effect on real income depends on whether
consumption of traded goods is higher or lower than yT
(whether there is a trade deficit or a trade surplus).
If there is a deficit, real income declines with a real
devaluation.
Reason: goods whose relative price has increased
have a higher weight in consumption than in income.
13



Demand for nontraded goods may increase because of
a higher level of output of traded goods.
Production of traded goods increases as long as the
price of its input does not rise by the full amount of the
devaluation.
Whether the latter condition holds depends on the
degree of wage indexation, the stance of inflationary
expectations, and other factors.
14
Effects Through Imported Inputs:
 Due to the presence of imported inputs, devaluation
may cause negative effect on the demand for
domestically produced goods.
 Reason: imported inputs make it more likely that the
real income effect of a devaluation is negative.
 To obtain national income imported inputs must be
subtracted from domestic output.
 Thus real devaluation affects real income also through
changes in the real value of imported inputs.
Two opposing effects of a real devaluation on the real
value of imported inputs:
 Real devaluation increases relative price of imported
inputs in terms of the basket of consumption; this
15
increases real value of initial volume of imported inputs.
If the price of labor does not increase by the full amount
of devaluation, relative price of imported inputs raises.
 Then domestic producers have an incentive to
substitute labor for imported inputs (reduce the
volume of imported inputs).
 Net effect depends on
 degree of factor substitutability in production;
 extent to which a devaluation is transmitted to
wages.
Assumptions:
 Traded goods are produced with a fixed amount of
specific capital and with labor.

16
Nontraded goods are produced with an imported input
and labor according to a CES production function with
elasticity of substitution .
Lizondo and Montiel (1989):
 (4) is modified in the presence of imported inputs by the
inclusion of an additional term:

z-JN[ - (1-)],

JN: volume of imported intermediate goods used in the
nontradable sector.
Net effect is ambiguous.
17

Net effect is more likely to be negative
 lower the elasticity of substitution between imported
inputs and primary factors;
 higher the share of nontraded goods in the price
index.
18
Income Redistribution Effects:
 Devaluation is the redistribution of income from sectors
with high propensity to spend on goods of this type to
sectors with a lower propensity.
Alexander (1952):
 Redistribution of income may affect expenditure, and
included it as one of the direct effects of devaluation on
absorption.
 Redistribution of income in two directions:
 from wages to profits because of lags in the
adjustment of wages to higher prices;
 from the private to the public sector because of the
existing structure of taxation.
19
Díaz Alejandro (1963); Krugman and Taylor (1978):
 Only impact effect of a devaluation is to redistribute a
given level of real income from wages to profits because
of an increase in prices (with constant nominal wages).
 This causes a reduction in the demand for domestic
output if marginal propensity to spend on home goods is
lower for profit recipients than for wage earners.
Example of other type of income redistribution effect
between workers and owners of capital:
 In a model with traded and nontraded goods, flexible
wages, and sector-specific capital, a real devaluation
 reduces real profits in the nontraded goods sector,
 increases real profits in the traded goods sector,
 has an ambiguous effect on real wages.
20
Cooper (1971):
 Possibility of redistribution from the factors engaged in
purely domestic industries to the factors engaged in
export- and import-competing industries.
 Although in some cases this may reduce demand, this
may also induce a spending boom.
 When all factors of production are mobile, redistribution
of income may depend on technological considerations.
 Example: in a Heckscher-Ohlin world, real wages and
profits depend on factor intensities.
 Real devaluation
 increases real payments to factors used intensively
by the traded goods sector;
 reduces real payments to the other factor.
21
How important is the effect on the demand for
domestic output of redistribution from wages to
profits?
Alexander (1952):
 What is important is the marginal propensity to spend.
 Even if profit recipients have a lower marginal
propensity to consume, higher profits may stimulate
investment.
 Thus redistribution of income may result in increased
absorption.
Díaz Alejandro (1963):
 Investment expenditure is more biased toward traded
goods than consumption expenditure.
22
Since investment expenditure is undertaken by profit
recipients, the demand for domestically produced goods
declines.
How important is the redistribution of income that will
lead to a change in the pattern of aggregate
expenditure?
 No firm support for the hypothesis of redistribution
against labor.
Edwards (1989b):
 In 15 out of 31 cases, there was no significant change
in income distribution.
 In 8 cases share of labor in GDP declined significantly.
 In 7 cases, it increased significantly.

23
Effects Through Changes in Real Tax Revenue:
 Effect of changes in real tax revenue operate through
 demand for domestic output
 private consumption expenditure or
 private investment;
 supply for domestic output.
 Many governments in developing countries derive a
substantial proportion of their revenues from import and
export taxes.
Krugman and Taylor (1978):
 Nominal devaluation that succeeds in depreciating the
real exchange rate will increase the real tax burden on
the private sector.
 Reason: real value of trade taxes increases.
24
This depends on the presence of ad valorem rather
than specific taxes on foreign trade.
Olivera-Tanzi effect:
 When lags in tax collection or delays in adjusting the
nominal value of specific taxes cause the real value of
tax collections to fall during periods of rising prices.
 If nominal devaluations increase inflation, the OliveraTanzi effect operates.
 Since real tax burden falls, devaluation would exert an
expansionary short-run effect on aggregate demand.
Third channel works through discretionary tax
changes:
 Assume, other than trade taxes, all taxes are levied on
households in lump-sum fashion.

25

Government's real tax receipts:
+ -
+
Tr = Tr(z, , ),

: parameter that captures the effects of discretionary
taxes;
: inflation rate.
First two terms in the function Tr(·) capture trade tax and
Olivera-Tanzi effects.
26

Government's budget constraint takes the form
Tr(z, , )
.
. (7)
 gN z - + gTz1 -  + i*z1 -  Fg – z1 -(Lg/E+Fg)

gT and gN: government spending on traded and
nontraded goods;
i*: foreign nominal interest rate;
Fg: net public external debt;
Lg: stock of net government liabilities to the central
bank.
Increase in Tr(·) must be offset somewhere else within
government budget, since (7) must hold at all times.
27





Effect of an increase in real trade taxes on aggregate
demand will depend on the nature of this offset.
Example: If the offset takes the form of a reduction in ,
leaving Tr(·) unchanged, contractionary effect on
aggregate demand disappears.
Nominal devaluation that results in a real depreciation
may affect each of the entries on the right-hand side of
Tr(·) identity.
External debt has been treated as if it were owed by the
private sector.
But most external debt has been owed by the public
sector.
28
Implications of public external debt:
 If the public sector is a net external debtor, a real
devaluation increases real value of interest payments
abroad.
 Government can finance increased debt service
payments by
 increased taxation,
 reduced spending,
 increased borrowing from the central bank or from
abroad.
 Effects on aggregate demand depend on the mode of
financing.
 If the government increases discretionary taxes, the
effects on aggregate demand would be contractionary.
29






Reason: private disposable income would fall. Reason:
private disposable income would fall.
If government reduces spending on nontraded goods,
contractionary effects on aggregate demand may
exceed those associated with tax financing.
If spending reductions fall on traded goods, the
contractionary effects would be nil.
If government finances by borrowing either from the
central bank or from abroad, contractionary effects fail
to appear.
Devaluation would affect the real value of government
expenditures on goods and services.
Total effect depends on the composition of government
spending between traded and nontraded goods.
30






Example: reduction in discretionary taxes may ensue
with corresponding expansionary effects on aggregate
demand.
Effect of a devaluation on discretionary taxes depends
on the monetary policy regime in effect.
This is captured by the last term on the right-hand side
of identity (7).
If the central bank pegs the flow of credit to the
government in nominal terms, the rise in prices
that
.
attends a nominal devaluation will reduce Lg/P.
This calls for an adjustment in the government budget,
possibly through a discretionary tax increase.
.
If the flow Lg is adjusted to accommodate the price
increase, no further changes in the budget will emanate
from this source.
31


If real valuation gains on the central bank's stock of
foreign exchange
.g reserves are passed along to the
government, L /P could increase.
Financing options would include an expansionary tax
reduction.
32
Wealth Effects:
 Increase in wealth is expected to increase household
consumption.
 Thus, devaluation can affect the demand for domestic
goods through its effects on real wealth.
 Nominal wealth is taken to coincide with the nominal
stock of money.
Alexander (1952):
 Devaluation would increase the price level and thus
reduce the real stock of money.
Two effects of this reduction:
 Direct effect, when individuals reduce their expenditures
to replenish their real money holdings to desired level.
33




Indirect effect, when individuals try to shift their
portfolios from other assets into money, thus driving up
domestic interest rate.
This contractionary effect on demand must be modified
if the private sector holds other types of assets whose
nominal value increases with a devaluation.
Assume that the private sector holds foreign-currencydenominated assets in an amount Fp.
Real wealth:
a = (M/P) + (EFp/P) = z1 - [(M/E) + Fp).
34

Percentage change in real wealth from nominal
devaluation:
a = (1-)z - ,
^
^
: share of domestic money in private sector wealth and
: devaluation rate.
Implications:
 If domestic money is the only asset,  = 1, devaluation
has a negative effect on real wealth and on demand.
 If the private sector also holds assets denominated in
foreign currency, the result is ambiguous.
35






Reason: although real value of domestic money
declines, real value of foreign assets increases if
domestic price level does not rise by the full amount of
devaluation.
Thus, effect on the demand for domestic goods may be
positive or negative.
It is more likely to be negative
 higher ;
^
 lower z;
 higher .
Presence of private sector external debt reduces Fp.
This increases .
Thus it increases the likelihood that a devaluation will
have a negative effect on real wealth.
36
Investment




Assume: capital stock in each sector consists of traded
and nontraded goods combined in fixed proportions.
A unit of capital in the traded goods sector consists of TN
units of nontraded goods and TT units of traded goods.
In the nontraded goods sector capital consists of NN
nontraded goods and TN traded goods.
Prices of a unit of capital in the traded goods sector PKT
and in the nontraded goods sector PKN:
PKT = TNPN + TTE,
(10)
PKN = NNPN + TNE.
(11)
37


Output in each sector is produced by using capital,
labor, and imported inputs.
Marginal product of capital in the two sectors:
- mTK = FKT (/E; KT),
- N
N
mK = FK(/PN, z; KN),



: nominal exchange rate.
In the In the short run, the capital stock is fixed.
By the first-order conditions for profit maximization, an
increase in wage will reduce demand for labor.
Ensuing increase in the capital intensity of production
will cause the marginal product of capital to fall.
38



Similar effect results from an increase in z.
Assume: all relative prices are expected to remain at
their post devaluation levels.
Sectoral net investment functions:
^T
K = qT
^N
K = qN
EmTK/PKT
{ i+-
KT
PNmTK/PKN
{ i+-
KN
}
-1
}
-1
Kh: rate of increase in the price of capital in sector h.
39




Net investment demand in each sector depends on the
ratio of the marginal product of capital to the real
interest rate.
Gross investment demand is the sum of net investment
and replacement investment.
Depletion is assumed to take place at the uniform rate
 > 0 in both sectors.
Combine (14) and (15) with replacement investment to
yield the total investment demand for nontraded goods:
IN =
^
T
N
T
IN + IN = NKTKT
+
(TNKT
+
NNKN)
+
^
N
NKNKN.
(16)
40
Branson (1986) and Buffie (1986b):
 Since substantial portion of any new investment in
developing countries consists of imported capital goods,
real depreciation raises price of capital in terms of home
goods.
 This discourages new investment and exerts
contractionary effect on aggregate demand.
 This is valid only in the case of investment demand that
originates in the nontraded goods sector.
 In traded goods sector, real depreciation lowers real
supply price of capital measured in terms of output.
 Thus, this effect stimulates investment in this sector.
 Thus, net effect on IN of changes in the supply price of
capital is ambiguous.
41
Second channel through which devaluation affects the
investment demand for nontraded goods operates
through real profits.
 Assume: firms operate on their factor demand curves.
 Return to capital is its marginal product, which depends
on
 initial stock of capital;
 product wage;
 real exchange rate in the case of the nontraded
goods sector.
van Wijnbergen (1986), Branson (1986), and Risager
(1988):
 First two contrasted the case of fixed nominal wages
with some degree of wage indexation.

42




Risager examined the effect on investment of holding
the nominal wage constant over some fixed initial
contract length and then restoring the initial real wage.
Result: devaluation may raise or lower the product wage
on impact depending on the nature and degree of wage
indexation.
With rigid nominal wages, the product wage would fall
on impact, and investment would increase in the short
run.
But with indexation, product wage could rise, thereby
dampening investment.
43





With some nominal wage flexibility, nominal devaluation
results in
 reduction in the product wage in the traded goods
sector;
 increase in the product wage in the nontraded goods
sector.
Investment would be stimulated in the former and
discouraged in the latter, with ambiguous effects on IN.
Third channel: in the presence of imported inputs.
Marginal product of capital in the nontraded goods
sector will be affected by a real devaluation through the
higher real costs of such inputs.
Effect is unambiguously contractionary.
44





Reason: depressing effect on profits in the nontraded
goods sector is not offset by positive effects on profits in
the sector producing traded goods.
In the case of a real depreciation that lowers the
product wage in the traded goods sector and raises it in
the nontraded goods sector, all three effects
 increase investment in the traded goods sector;
 decrease it in the nontraded goods sector.
If these effects are sufficiently strong, IN must increase
when capital is sector specific.
In this case, increase in investment demand in traded
goods sector can be met only through new production.
It cannot be offset by negative gross investment in the
nontraded goods sector.
45
Nominal Interest Rates


Increase in real interest rate reduces
 private consumption of nontraded goods;
 investment spending on nontraded goods by both the
traded and nontraded goods sectors.
Distinguish between
 current effect of an anticipated devaluation;
 contemporaneous effect of a previously unanticipated
devaluation.
46
Assumptions:
 Domestic residents can hold financial assets in the form
of money, domestic interest-bearing assets, and
interest-bearing claims on foreigners.
 Domestic interest-bearing assets take the form of loans
extended by households to other entities in the private
sector.
 Portfolio adjustment is costless.
 Effects of a devaluation on the nominal interest rate
charged on these loans depend on
 degree of capital mobility;
 severity of portfolio adjustment costs.
 Distinguish two cases based on whether domestic loans
and foreign assets are perfect or imperfect substitutes.
47

If loans and foreign assets are imperfect substitutes,
equilibrium in the loan market:
- +
+ M + EFp +
h i, i* + a, y,
;x
P
(
) = 0,
(17)
h(·): real excess demand function for loans;
i: nominal interest rate on loans;
i* + a: nominal rate of return on foreign assets;
i*: foreign nominal interest rate;
a: expected rate of depreciation of domestic currency;
y: real income;
(M+EFp)/P: real household financial wealth;
x: vector of additional variables.
48





Increase in i has a negative own-price effect on excess
loan demand.
Increase in i* + a raises excess demand for loans as
borrowers switch to domestic finance while lenders seek
to place more of their funds in foreign assets.
Increase in y causes lenders to increase their demand
for money, thereby increasing excess demand in the
loan market.
Increase in (M+EFp)/P both reduces borrowers‘ need for
outside financing and provides lenders with surplus
funds.
This effect reduces excess demand in the loan market.
49
Effect of a devaluation on i at a given initial level of y
and PN with a = 0:
 Unanticipated devaluation: effect on i depends on the
composition of household financial wealth.
 Whether real excess demand for loans rises or falls
depends on whether (M+EFp)/P increases or decreases.
 Devaluation lowers real money stock but raises real
value of foreign assets.
 Former effect dominates if
 large share of household financial wealth is devoted
to cash balances;
 traded goods have a large weight in private
consumption.
50
In this case, (M+EFp)/P falls, real excess demand for
loans increases, and i rises.
 This result is reversed, if
 foreign assets dominate households' balance sheets;
 traded goods carry a small weight in domestic
consumption.
van Wijnbergen (1986):
 Households hold no foreign assets.
 Thus, nominal devaluation raises i.
Buffie (1984a):
 Households hold a substantial portion of their wealth in
assets denominated in foreign exchange.
 Thus, nominal devaluation reduces i.

51


When the partial derivative hi, evaluated at i* + a,
approaches negative infinity, domestic loans and foreign
assets become perfect substitutes.
(17) is replaced by
i = i* + a.
Thus, uncovered interest parity holds continuously.
 Under these conditions, unanticipated current
devaluation has no effect on i.
Effects of anticipated future devaluation.
Imperfect substitutability:
a
 It is represented by an increase in  in (17), with the
level of the exchange rate held constant.
 Thus, i rises.
52


If the own-price effect hi exceeds the cross-price effect
hi* +  , increase is lower than the anticipated devaluation.
Perfect substitutability:
i rises by full amount of the anticipated devaluation.
New structuralist school emphasizes the importance of
“working capital” in developing countries as a source of
loan demand.
This introduces effects of unanticipated current
devaluation.
a



53

These effects can be captured by defining x:
+ + x = x(w, E, PN).




x: index of real working capital requirements depend on
 wage bill;
 purchases of imported inputs.
Increase in x, increases demand for loans.
Real working capital requirements increases when
 nominal wage;
 domestic-currency price of traded goods increases.
It falls when the price of nontraded goods rises.
54



Because unanticipated current devaluation increases w,
real excess demand for loans rises, putting upward
pressure on i.
Taking working capital into account causes the impact
on i to be positive even if foreign assets are prominent
in private sector balance sheets.
Thus, working capital considerations enhance the
likelihood that devaluation will be contractionary.
55
Effects on Aggregate Supply


Production cost of goods in domestic currency
increases as the prices of the factors of production rise
in response to a devaluation.
Devaluation causes upward shift in supply curve
through three separate channels:
 increases in nominal wages;
 use of imported inputs;
 increases in the cost of working capital.
56
Effects on Nominal Wage


Assume “dependent economy” setup, take capital to be
sector specific and fixed in the short run, and allow both
sectors to employ imported inputs.
Aggregate demand for labor:
nd = n0 - d1(w-e) - d2(w-pN) - d3(e-pN),
= n - (d1 + d2)(w - e) - (d2 + d3)z,
(20)
n0, d1, d2, and d3 are positive parameters.
57



Increase in the product wage measured in terms of
traded goods reduces the demand for labor in the
traded goods sector by
 reducing output in that sector;
 encouraging substitution of imported inputs for labor.
Magnitude of d1 depends on
 share of labor employed in the traded goods sector;
 labor intensity of production in that sector;
 elasticity of substitution between labor and imported
inputs in that sector.
Sign and magnitude of d2 are determined similarly but
for nontraded goods sector.
58


d3: effect on demand for labor in nontraded goods
sector of an increase in price of imported inputs;
 demand for labor falls because of a decrease in the
level of output;
 it increases as labor is substituted for imported
inputs.
Magnitude of d3 depends on
 substitution elasticity;
 labor intensity of output in nontraded goods;
 share of the labor force employed in that sector.
59

Aggregate supply: current nominal wage is given by
~
w=w
+ s1(n–n0) + s2pa + s3(p–pa)
~
= w + s1(n–n0) + s3e – s3(1-)e
(21)
+ (s2–s3)[ea – (1-)za],
~ s , s , and s are positive parameters.
w,
1
2
3
60
Wages depends on
 n: employment;
 n0: “natural” or full-employment level;
 price expectations for the contract period formed
when the contract was signed;
 degree of indexation s3 to unanticipated price shocks
(p - pa).
Special cases derived from Equation (21):
 Exogenous nominal wages follow from s1 = s2 = s3 = 0.
 Predetermined nominal wages with Fischer-type
contracts are implied by s2 = 1 and s1 = s3 =0.
 Wage indexation to the current price level is imposed by
setting s1 = 0 and s2 = s3. As a special case, fixed real
wages follow from s1 = 0 and s2 = s3 = 1.
61





Phillips curve is derived with s2 = s3 = 0.
Neoclassical labor market model can be produced by
setting s2 = s3 = 1.
Friedman-Phelps version of the Phillips curve emerges
from s2 = 1 and s3 = 0.
Impose only s2 = 1 and s3 < 1 restrictions so that
 perfectly anticipated inflation has no effect on
workers’ real wage demands;
 degree of indexation to current prices is only partial.
62

Substituting (20) in (21), equilibrium nominal wage:
w=
ea
1 -  + 23 a s3 + 12 z +
(e – ea)
1 + 12
1 + 12 (22)
s3(1-) + 23
(z-za),
1 + 12
12 = s1(d1+d2), and 23 = s1(d2+d3).
Observations in assessing effects on the nominal
wage of an exchange-rate depreciation:
 How nominal depreciation translates into a real
depreciation is crucial.
63




In the absence of perfect indexation, it is important to
distinguish whether devaluation was previously
anticipated or not.
In neither case nominal wage necessarily increases.
Assumption: price of nontraded goods is constant on
impact.
(22) can be written as
w=
ea
 + s1(d1–d3) a
e +
1 + 12
s3 + s1(d1–d3)
(e – ea)
1 + 12
(23)
64





If d3 > d1, effects of both an anticipated and an
unanticipated devaluation could be negative.
Increase in E will lower demand for labor, given wages
and the price of nontraded goods.
Reason: increase in demand in the traded goods sector
is offset by reduced demand in the nontraded goods
sector.
The latter arises from an increase in the price of
imported inputs, which reduces the level of output and
demand for labor.
This effect is dominant if
 share of labor in the nontraded goods sector is large;
 this sector is intensive in its use of imported inputs;
65
elasticity of substitution of labor for imported inputs in
that sector is small.
Imported inputs in nontraded goods sector dampens
increase in w that would accompany devaluation.
This acts as an offset to the contractionary effect of a
devaluation on the supply of nontraded goods.
If d1 > d3, as long as a nominal depreciation results in a
less-than-proportional real depreciation, increase in w is
 no greater than increase in PT;
 no less than increase in PN.
Product wage falls in traded goods sector and rises in
nontraded goods sector.





66
Imported Inputs
Devaluation increases price of imported inputs by the
same percentage as the exchange rate.
 This drives up costs of production of goods.
 Magnitude of this increase in costs depends on
 technological factors;
 extent to which the price of other factors of
production responds to the devaluation.
Assumptions:
 Economy produces and consumes traded and
nontraded goods.
 Nontraded goods are produced with imported inputs
and “value added,” according to CES production
67
function with elasticity of substitution .






Share of labor in value added is denoted by .
Nominal wages are determined exogenously and to
increase by devaluation.
Return on capital is endogenous and varies to clear the
market for that factor.
Effect of devaluation on the supply of nontraded goods
is investigated by an increase in supply price.
Percentage increase in the supply price
^ +  r,
^
N = J + ww
k
(24)
^ percentage of the nominal devaluation;
:
^ exogenous increase in nominal wages;
w:
r: endogenous increase in the return of capital.
68

Since labor and capital are combined according to a
Cobb-Douglas production function and capital is
constant,
^ + n.
^
r^ = w

Cost minimization implies
^
n^ = J{w + J[(1-) + ]}-1(-w),
^ +  { +  [(1-) + ]}-1(-w).
^
r^ = w
J w
J


(25)
(25) is useful for examining effect of devaluation and
adjustment of wages on r.
^ = , r increases by the same amount.
If w
69
Reason:
 There is an incentive to substitute value added for
imported inputs, and within value added to substitute
capital for labor.
 Since capital is constant, r increases until initial ratio of
w to r is restored, so that initial desired capital/labor
ratio is also restored.
^ ^
 At the end, r = w = , and same combination of inputs is
used.
^
^
 r different from w would not be an equilibrium value.
 If w does not increase by the full amount of the
devaluation, r increases by more than w.
70
Reason:
 If r increases only by the same amount as w, producers
wants to use the same capital/labor ratio.
 Since capital is fixed, this implies constant employment.
 But since price of value added would decline relative to
price of imported inputs, there would be an excess
demand for capital and labor.
 These demands are satisfied by an increase in the use
of labor and in r.
^ in (24) yields
 Using (25) to replace r
^
N =  - (1-J){(1-J) + (1-) + }-1(-w).
71




If w increase by the full amount of the devaluation,
supply curve shifts upward by the same percentage as
the exchange rate.
If w does not rise by the full amount of the devaluation,
supply curve shifts upward by less than the exchange
rate, but by more than increase in w.
Increase in supply price will be larger
 larger is the share of imported inputs in total costs;
 larger is the share of capital in value added.
Increase in supply price will be larger the smaller is
elasticity of substitution between imported inputs and
value added.
72


If CES function were assumed for the production of
^ < , increase in supply price would
value added, when w
be larger the lower the elasticity of substitution between
labor and capital.
If w increase by less than the full amount of the
devaluation, output of traded goods will increase.
73
Effects Through Costs of Working Capital
Contractionary effects of nominal devaluation on supply
of domestic output work by increasing cost of working
capital.
 Consider nontraded goods sector.
 Need to finance working capital arises from asynchrony
between payments and receipts.
Assumptions:
 To finance real wage bill NnN (N  w/PN is product
wage in nontraded good sector) and real imported input
bill zON, firms hold real stocks of loans outstanding in
n
 h (i, wNnN) for real wages;
o
 h (i, zON) for imported inputs.

74

Representative firm's profits:
N = PNyN(nN, ON) - nN – EON – iPNhn(·)
+ iPNho(·).

First order conditions for profit maximization:
dyN/dnN = N[1 + ihn n (·)],
N N
dyN/dON = z[1 + iho n (·)].
N N
75

These equations can be solved for labor and imported
input demand functions:
- ? nN = nN(N, z, i),
? - d
d
ON = ON(N, z, i).
d

d
Substituting them in short-run production function for
nontraded goods yields short-run supply function for
nontraded goods:
- yN = yN(N, z, i).
s
s
(32)
76

Traded-goods supply function:
- yT = yT(T, i),
s
s
(33)
where T = w/E.
 Presence of costs of financing working capital has two
supply consequences that affect the likelihood of
contractionary devaluation:
 Cavallo-Patman effect;
 effect of working capital costs on the elasticities of
(32) and (33) curves.
Cavallo-Patman effect:
 Increase in loan interest rates
 adds to the costs of financing working capital;
77
shifts the output supply curve upward.
This effect is captured in (-) sign of the partial derivative
of i in (32) and (33).
Magnitude of the effect depends on properties of the
functions hn and ho.
Important aspects of this effect:
 Appears in conjunction with unanticipated
devaluation only if capital mobility is imperfect.
 If domestic interest rates rise, then this effect is the
only channel through which devaluation may exert
contractionary effects in traded goods sector.
 This effect represents a second channel through
which anticipated future devaluation could affect
current output.




78
Effect of working capital costs on the elasticities of
(32) and (33) curves.
 This is captured by cross-partial derivatives of these
equations.
 Working capital costs are likely to reduce short-run
supply elasticities in both sectors.
 Reason: increase in marginal costs associated with the
need to finance additional working capital.
 Real exchange-rate depreciation causes this reduction
in elasticities to be unfavorable in traded goods sector.
 But the reduction may be either favorable or
unfavorable in nontraded goods sector depending on
whether demand for such goods contracts or expands.
79
Empirical Evidence
Four alternative empirical approaches to analyze
contractionary effects of devaluation:
 Factual approach: changes in output performance at the
time of devaluation( “before-after” methodology).
 “Comparison” or “control group” approach: compares
output growth in devaluing countries with performance
in a group of nondevaluing countries.
 Applying econometric methods to time series data in
order to quantify the impact of exchange-rate changes
on real output.
 Using simulation models or reduced-form equations to
analyze effects of exchange-rate variables on output.
80
Before-After Approach
Díaz Alejandro's (1965):
 Experience of Argentina over the period 1955-1961.
 1959 devaluation of the peso was contractionary.
 Reason: shift in income distribution toward high savers,
which depressed consumption and real absorption.
Cooper (1971):
 24 devaluation episodes between 1959 and 1966.
 Assessed response of balance of trade and payments,
inflation, and elements of aggregate demand.
 Trade balance and balance of payments improved in
most cases.
 Evidence of contractionary effects following devaluation.
81
Two weaknesses:
 examined only short-run effects of devaluation;
 his approach was based on a strict assumption.
Killick, Malik, and Manuel (1992):
 Reviewed the results of 266 IMF-supported programs
implemented during the 1980s.
 Many of them incorporated nominal devaluation as a
key policy measure.
 Although programs have no discernible effect on output
growth in short term, over longer term growth rates
improve.
 These programs were associated with substantial fall in
share of investment in output.

82
Control Group Approach
Overcomes inability of before-after approach by
distinguishing between
 effect of devaluation per se;
 effect of other factors on output.
 Basic assumption: devaluing and nondevaluing
countries face the same external environment.
 Thus, difference in group performance reflects only
effect of exchange-rate changes.
Kamin (1988):
 50 to 90 devaluations out of 107 devaluations between
1953 and 1983.

83
Statistical tests for significance of changes over time for
each economic indicator for both
 devaluing country's performance and;
 comparison group's performance.
 Growth rate remains in general positive.
 Sharp and significant decline in output growth is seen
in devaluing countries, but it occurs in the year
preceding devaluation.
 Growth in subsequent years turns upward and improves
relative to comparison group.
Edwards (1989b):
 Evolution of a number of key variables during 3 years
preceding and 3 years following 18 devaluations
episodes between 1962 and 1982 in Latin America.

84
Two groups:
 control group used fixed nominal exchange rate;
 devaluing countries.
 Nonparametric statistical tests are used in comparisons.
 Decline in output growth in periods surrounding
devaluations may not be a consequence of exchangerate changes.
 But it may reflect the imposition of trade and capital
restrictions that have accompanied devaluations.
Donovan (1981):
 12 IMF-supported devaluations between 1970 and
1976.
 Compares performance of devaluing economies with
that of all non-oil-exporting developing countries. 85

Reductions in GDP growth were registered only for
those programs aimed at import restraint.
 However, in a subsequent study
Donovan (1982):
 Sample of IMF-supported programs was extended.
 Economic growth fell by more than average decline
experienced by non-oil developing countries in the oneyear comparisons.
 But it fell by less in the three-year comparisons.
Gylfason (1987):
 32 IMF-supported programs during 1977-1979.
 Differences in output growth between countries with IMF
programs and non program countries were not
statistically significant.
86

Khan (1990):
 Effects of IMF-supported programs on balance of
payments, current account, inflation, and growth in a
group of 69 developing countries over 1973-1988.
 Real exchange-rate changes have negative effect on
rate of growth of output, but the coefficient is small and
not highly significant.
87
Econometric Models
Sheehey (1986):
 Uses Lucas-type supply function and cross-country data
for 16 Latin American countries to estimate impact on
short-run output growth of
 unanticipated inflation
 changes in relative cost of foreign exchange;
 business cycle fluctuations in industrial countries.
 Support the contractionary devaluation hypothesis.
 Strong impact of external economic activity on real
growth rates.
Edwards (1986):
 Data on 12 developing countries for 1965-1980.
88
Estimates model of real output behavior.
 Real exchange-rate changes have small contractionary
effect in the short run.
 In the medium run, perverse effect is reversed.
 In the long run devaluations are neutral.
Edwards (1989b):
 Multisector macroeconomic model of a dependent
economy with imported intermediate goods, foreign
debt, and wage indexation.
 Analyzes how devaluations affect aggregate output and
employment.
 Devaluations have a short-run contractionary effect on
real output.
 Long-run neutrality result was found to hold again.
89

Weakness:
 Econometric tests make no distinction between
anticipated and unanticipated changes in key
macroeconomic variables.
 In models when there is no distinction between
anticipated and unanticipated movements, impact of
devaluation is theoretically ambiguous.
 This result derives from two conflicting factors:
 devaluation generates an expansionary effect
through aggregate demand;
 through its effect on cost of imported intermediate
inputs, negative impact on aggregate supply.
 So devaluation can be contractionary even if net effect
on aggregate demand is expansionary.
90
Agénor (1991):
 Makes this distinction in his model.
 Impact of real exchange-rate devaluation on real output
is different.
Anticipated depreciation of real exchange rate:
 This causes a rise in expected price level.
 Workers thus increase their nominal wage demands.
 Demand for both labor and imported inputs falls, and
output also falls.
Unanticipated devaluation:
 It has no effect on price expectations or the real wage.
 It causes unanticipated increase in domestic demand as
relative price of domestic output falls.
91
This implies an unanticipated increase in prices, which
stimulates aggregate supply.
 Econometric tests shows:
 anticipated depreciation of the real exchange rate
has a negative effect on economic activity;
 unanticipated depreciation has a positive impact;
 contractionary effect of anticipated depreciations
remains significant even after a year.
Kamin (1988):
 Time series regression analysis of exchange-rate
effects on output may not be appropriate for
characterizing devaluation episodes.
 They do not tell us what happened during a specific
devaluation episode.

92


Devaluations associated with other stabilization policies,
but they are also large, isolated events that occur only
sporadically.
Annual models may not use proper time frame to
examine long-run effects of devaluations.
93
Macro-Simulation Studies
Gylfason and Schmid (1983):
 Log-linear macro model of an open economy with
intermediate goods.
 In their model, devaluation exerts conflicting effects:
 it has an expansionary effect through aggregate
demand;
 it has a negative effect on aggregate supply.
 Empirical results: devaluation is expansionary in 8 out of
10 countries.
Gylfason and Risager (1984):
 Model stresses effects of exchange-rate changes on
interest payments on external debt.
94
While devaluation is expansionary in developed
economies, it is contractionary in developing countries.
Solimano (1986):
 Uses computable macroeconomic model for Chile.
 Devaluation is contractionary in short to medium run.
 Focused on three factors:
 structure of trade sector in terms of response of trade
flows to changes in relative prices;
 relative intensity of domestic value added with
respect to imported inputs in production across
export- and import-competing industries;
 degree of wage indexing.
Branson (1986):
 Stagflationary effect of devaluation for Kenya.
95

Uses two-sector model with sticky prices, wage
indexing, and imported intermediate goods.
Roca and Priale (1987):
 Peru during which the Peruvian authorities attempted to
reduce current account deficit by depreciating real
exchange rate through large nominal devaluations.
 Since large portion of loans to the business sector were
contracted in U.S. dollars, devaluations
 increased price of imported inputs;
 raised real cost of credit, pushing up the cost of
working capital for highly indebted firms and
generating a strong stagflationary effect.
Kamas (1992):
 Estimates a model for Colombia.

96
Distingished traditional and nontraditional exports.
 Nominal devaluation is more likely to be contractionary
 lower elasticity of substitution between capital and
labor;
 lower the elasticity of substitution between imported
inputs and domestic value added;
 higher degree of wage indexation.
Limitations of before-after approach:
 Not take into account other variables such as monetary
and fiscal policies, external disturbances, and structural
changes.
 Focusing on “before” and “after” makes it harder to
detect causality among variables.

97
Limitations of control group approach:
 Devaluing countries differfrom non-devaluing countries
prior to a devaluation episode.
 This matters in evaluating impact of a currency change
on output.
Limitations of simulation approach:
 Use imputed parameter values.
 This causes questions about reliability of results derived
from a set of “guesstimates” and coefficients that are
not consistently estimated.
Limitations of econometric studies:
 Specification of output equation is often arbitrary.
 It does not always provide basis for distinguishing
anticipated and unanticipated movements in variables.
98
Real-Exchange-Rate
Targeting



The Analytical Framework.
Targeting the Real Exchange Rate.
Effects of Macroeconomic Shocks.
100
The Analytical Framework






Small open economy in which competitive firms
combine homogeneous labor and sector-specific capital
to produce nontraded goods and traded goods.
To allow the consideration of terms-of-trade shocks,
three-good framework is adopted by separating traded
goods into exportables and importables.
Former is produced at home and the latter is produced
abroad.
All prices are flexible, ensuring that full employment is
continuously maintained.
Income generated from production of the two goods is
received by consumers.
They use it to buy home goods and importables. 101




Households allocate a constant fraction of their total
expenditures to each of the two goods in every period.
Real value of aggregate household expenditures
depends on
 real value of factor income net of taxes;
 real interest rate;
 real financial wealth.
Real factor income consists of the real value of output
of exportables and home goods measured in terms of
consumption basket.
It is an increasing function of the terms of trade, so it
can be writen as y = y(), where  is terms of trade.
102



Real household financial wealth in terms of importables
(a) consists of
 real money balances m;
p
 plus real value of foreign securities F ;
 minus real value of loans extended by the banking
system to households Lp.
za: real financial wealth.
 z: relative price of importables in terms of nontraded
goods.
 : share of nontradables in the consumption basket.

 z : relative price of importables in terms of the
consumption basket.
Capital is perfectly mobile, which ensures that nominal
interest rate parity is continuously maintained.
103



Under fixed nominal exchange rates, this implies that
domestic nominal interest rate (real rate r plus rate of
inflation ) is equal to the foreign interest rate.
^ domestic real interest rate.
i* + z:
^
 z: expected and actual rate of change of the real
exchange rate.
 i*: foreign interest rate.
Equilibrium can be analyzed from two relationships:
 describing private wealth accumulation;
 equilibrium in the market for nontraded goods.
104

Private wealth accumulation measured in terms of
importables is equal to
 real factor income net of real lump-sum taxes y - ;
 less household consumption spending c;
 plus real interest earnings on existing holdings of
nonmonetary financial assets;
 less revenue from inflation tax that accrues to the
central bank:
.
^ 
a = y -  - c(y - , i* + z,
z a) + i*a
(34)
– (i*+)md(y, i* + ),
md(·): real demand for money;
: devaluation rate.
105





Government is assumed to consume the same two
goods and to finance its expenditures
 by levying taxes, through the receipt of transfers from
the central bank;
 by borrowing.
Government budget must satisfy the standard
intertemporal solvency constraint.
Shocks affect government budget through transfers
from the central bank.
Change in level of these transfers gives rise to one-forone changes in government spending on importable
goods.
Thus government's budget is kept in balance.
106




Current account of the balance of payments will be
equal to private saving.
Under
. fixed exchange rates, private saving will be equal
to a.
So (34) is external balance condition.
Because prices are flexible, nontraded goods market
must clear continuously:
s
yN(, z) =

z1 - c(y
^
- , i* + z, za) + gN, (35)
yNs(·): supply function for nontraded goods;
gN: exogenous level of government consumption of
such goods.
(35): internal balance condition.
107




(34) and (35) together determine paths of real exchange
rate and real wealth.
Economy approaches a long-run equilibrium when both
z and a reach a constant value.
Since real exchange rate is constant in the fixed
exchange-rate steady state, domestic inflation rate must
equal the world inflation rate (which is zero).
.
^
Since z and a are both zero, under nominal exchange
rate targeting (34) and (35) jointly determine steadystate equilibrium values of real exchange rate and real
wealth.
108
Targeting the Real
Exchange Rate





Authorities choose as their real-exchange-rate target
that value of z that corresponds to steady-state
equilibrium under fixed exchange rates.
In (34),  is no longer zero, but must be replaced by the
domestic inflation rate .
z becomes a constant (can be set equal to unity).
Although nominal private wealth is predetermined, real
wealth can change in the short run, through changes in
domestic price level.
Thus, a becomes an endogenous variable.
109



So internal balance is maintained by changes in the
aggregate price level.
As long as exogenous and policy variables that affect
.
the market for nontraded goods do not change, a = 0
must hold continuously.
Under these circumstances, (34) and (35) become:
y() -  - c[y()-, a; i*] + i*a – (i*+)md(·) = 0,
(36)
yN() - c[y()-, a; i*] – gN = 0.
(37)

Any nominal wealth accumulation by private sector
must be offset by price level increase.
110






Implication: real shocks have consequences for the
equilibrium domestic inflation rate under real-exchangerate targeting.
Shock raises domestic price level, thereby lowering a.
From (34), this reduces consumption and increase
.
private saving, causing a to become positive.
This causes private consumption to increase over time,
leading to an excess demand for nontraded goods.
In order to maintain equilibrium in the home goods
market, domestic prices must rise to maintain real
wealth continuously at its new equilibrium level.
Inflation tax must be high enough to induce private
agents to willingly hold stock of real wealth necessary to
clear the market for nontraded goods.
111









Figure 8.1: determination of equilibrium.
Initial level of real exchange
rate is assumed to satisfy
.
(34) and (35) with z^ and a both zero.
When world inflation is zero, domestic inflation is also
zero.
Vertical axis: domestic rate of inflation, .
Horizontal axis: level of real wealth.
(37) is represented by NN.
This shows that there is only one level of real assets
that will clear the market for home goods.
(36) is represented by AA.
“C” shape of this locus reflects the fact that,
 rise in  increases inflation tax for low  but,
 reduces it once  becomes sufficiently high.
112
F
i
g
u
r
e
8
.
1
E
q
u
i
l
i
b
r
i
u
m
u
n
d
e
r
a
R
e
a
l
E
x
c
h
a
n
g
e
R
a
t
e
R
u
l
e
*

N
A
E
N
a
A
S
o
u
r
c
e
:
M
o
n
t
i
e
l
a
n
d
O
s
t
r
y
(
1
9
9
1
,
p
.
8
8
1
)
.
113




.
.
If a > 0, increase in  will be required to reduce a at low
rates of inflation.
But reduction in  will be required for values of  that
are sufficiently high.
Assume government (adopting real-exchange-rate
target), remains at the low-inflation equilibrium (E).
This corresponds to long-run equilibrium under fixed
exchange rates.
114
Effects of Macroeconomic
Shocks
Effects of an improvement in terms of trade (increase
in ):
 This raises both real factor income and consumption.
 But, under the assumption of positive marginal
propensity to save, this leads to
 increase in private saving;
 hence accumulation of wealth.
 Elimination of this rise requires rise in a to satisfy the
.
condition a = 0.
 Figure 8.2: AA curve shifts to the right in response to an
improvement in the terms of trade.
115
F
i
g
u
r
e
8
.
2
E
f
f
e
c
t
o
f
a
n
I
m
p
r
o
v
e
m
e
n
t
i
n
t
h
e
T
e
r
m
s
o
f
T
r
a
d
e

N
'
N
A
A
'
E
'
A
'
E
N
'
N
a
A
S
o
u
r
c
e
:
M
o
n
t
i
e
l
a
n
d
O
s
t
r
y
(
1
9
9
1
,
p
.
8
8
6
)
.
116






Improvement in terms of trade raises real product wage
in the home goods sector and causes reduction in the
supply of nontradable goods.
Real factor income rises, which increases demand for
home goods.
Thus, both supply and demand effects lead to excess
demand in home goods market.
This requires reduction in a to restore market
equilibrium.
Thus, NN shifts leftward in Figure 8.2, to N’N’.
From Figure 8.2, at the original level of inflation, real
private wealth would be increasing.
117





.
Since a is positive, this generates excess demand
pressures in the home goods market, putting upward
pressure on domestic prices.
Equilibrium is reached only once inflation and inflation
tax have increased to a level sufficient to induce the
private sector to hold the new equilibrium value of a.
New steady-state position of the economy is at E’.
Policy of fixing nominal credit stock would have no
effect on domestic inflation.
Reason: control of credit stock does not give authorities
a handle on private spending under perfect capital
mobility.
118






With domestic credit and foreign assets being perfect
substitutes, a reduction in stock of credit would be offset
by private sector through reduction of its rate of
accumulation of foreign assets.
Economy's equilibrium would also be unchanged.
Income of neither sector would be affected.
Reason: central bank would collect directly from
foreigners the interest income that was previously
passed on to it by the private sector.
Thus, nominal credit target affects only the capital
account of the balance of payments.
It leaves the rate of inflation unchanged.
119