Transcript Monetary Policy in Post-Stabilization LICs

The Macroeconomics of Managing
Increased Aid: Country Experiences
“Global Conference on Gearing Macroeconomic
Policies to Reverse the HIV/AIDS Epidemic”
November 2006
Jan Kees Martijn
International Monetary Fund
Outline

A framework for assessing macro
implications of increasing aid:





absorption and spending
Findings from case studies
Prescription under IMF-supported programs
Related evidence on monetary programs
Concluding thoughts
Defining Spending and Absorption

Spending of aid: widening of the fiscal deficit,
net of aid, as a result of new aid.
–

Spending = (Govt. expenditures-Domestic
revenues)/Aid
Absorption: widening of the current account
deficit, net of aid, in response to an increase in
aid.
–
–
Absorption = (non-aid current account
deficit)/Aid
Central bank’s willingness to sell foreign exchange
is crucial.
Policy Responses – Four Options
1. Absorbed and Spent


Textbook case where central bank sells
aid foreign exchange and fiscal deficit
rises as aid is spent.
No change in money supply. Risks Dutch
disease.
3. Not Absorbed but Spent
–
–
–
Central bank accumulates foreign exchange
as reserves; fiscal deficit rises as aid is spent.
No real resource transfer.
Unsterilized: Money supply rises. Risks
inflation.
Sterilized: Crowding out of private sector.
Costly domestic debt.
2. Absorbed but not Spent
–
–
Central bank sells foreign exchange but
fiscal deficit remains unchanged.
Helps achieve stabilization, lower debt,
provides resources for private
investment.
4. Not Absorbed, not Spent
–
–
Central bank accumulates foreign
exchange as reserves; fiscal deficit
net of aid unchanged. No real
resource transfer.
Equivalent to rejecting aid (in long
run).
Main Findings (1)

No evidence of Dutch disease—real exchange
rates did not appreciate.
Real Effective Exchange Rate
140
120
100
80
60
t=-1
t=0
t=1
Aid surge period
Ghana
T anzania
reference line=100
t=2
Ethiopia
Mozambique
Uganda
t=3
Main Findings (2)

Aid was not fully used—in no case was it both
spent and absorbed.
Was Aid Spent?
8
8
6
in percent GDP
12
4
0
-4
-8
4
2
0
Increase in Current Account Deficit Net
of Aid
Et
na
hi
op
M
ia
oz
am
bi
qu
e
Ta
nz
an
ia
Ug
an
da
Increase in Aid
-2
Gh
a
hi
op
M
ia
oz
am
bi
qu
e
Ta
nz
an
ia
Ug
an
da
Et
na
-12
Gh
a
in percent GDP
Was Aid Absorbed?
Increase in Budgetary Aid
Increase in Fiscal Deficit Net of Aid
Main Findings (3)

Some countries neither spent nor absorbed: aid
went into reserves and spending did not
increase.
–
–

Ghana, Ethiopia
Why? Desire to build up reserves or smooth volatile
aid flows
Some countries spent the aid but resisted
absorption. This amounts to domestic
financing of spending.
–
–
Most common, though unattractive, response:
Tanzania, Uganda, Mozambique.
Why this choice? One factor: fear of appreciation.
What Did the IMF Program Prescribe?

PRGF programs generally encouraged an absorb-andspend policy.
– Fiscal deficit net of aid incorporated planned
increases; reserve targets were modest.
– In a few cases, absorption without spending was
envisaged when macro stability not established or
domestic debt too high initially (e.g. Ghana)

However, PRGF programs often dealt with
aid surprises more cautiously. Asymmetric
adjusters with respect to aid surprises
common.
Lesson: the Need for Coordinated fiscal and
Monetary Policies

Competing objectives. Suppose the Minister of Finance
(and donors) care mostly about increasing spending
and the central bank mainly dislikes appreciation.

Coordinated choice. Either #1 or
#4, depending on benefits of
higher govt. spending vs.
adverse impact of appreciation.

Uncoordinated outcome.
Suppose also that the Minister
controls spending and the CB
controls absorption. This could
lead to suboptimal outcome (#3).
1. Spend and
Absorb
2. Absorb
but not
Spend
3. Spend but
not
Absorb
4. Neither
spend
nor
absorb
Performance under Monetary Programs


Corresponding performance under monetary
programs: reserves higher and domestic credit
lower than programmed under IMF programs.
(based on 1999-2004 performance of 15 Poststabilization LICs)
Background: The Monetary Impact of Aid




When the government receives
aid, it sells foreign exchange to
the central bank and gets local
currency deposits.
ΔNFA=-(ΔNDA), no effect on
reserve money.
Aid spent → govt. draws down
deposits, NDA and money
increase.
Now CB faces a choice: sell
foreign exchange, sterilize, or
allow inflation.
Aid inflows and money supply
Central Bank Balance Sheet
NFA
+100 RM
0
NDA
-100
Central Bank Balance Sheet
NFA
+100 RM
+100
NDA
0
Monetary Projections and Projection Deviations
(averages, in percent)
t-2
Projection as of 1/
t-1
t(1)
Outcome
t(2)
Projected
Inflation (end year CPI)
Inflation (GDP deflator)
Real GDP growth
Broad money growth
Reserve money growth
NDA contribution
NFA contribution
Velocity (% change)
Money multiplier (% change)
# observations
4.1
4.3
6.0
12.0
4.9
5.2
5.9
12.3
10.5
-0.6
10.3
-0.6
1.6
6.0
6.3
4.9
12.2
9.7
-6.8
16.4
-0.4
2.7
5.9
6.1
4.9
13.3
10.3
-7.8
18.2
-1.6
3.1
6.3
6.8
5.1
17.5
15.7
-11.5
27.1
-4.3
2.4
29
43
54
54
54
-0.7
-1.0
0.2
-4.5 **
-0.5
-1.0
0.6
-5.9 ***
-4.6 *
6.9
-13.1
4.5 ***
-1.4
-0.3
-0.5
-0.2
-5.4
-6.0
5.1
-11.3
3.8
0.2
-0.4
-0.7
-0.2
-4.3
-5.4
3.7
-8.9
2.7
0.7
Deviations from projections 3/ 4/
Inflation (end year CPI)
Inflation (GDP deflator)
Real GDP growth
Broad money growth
Reserve money growth
NDA contributio
NFA contributio
Velocity (% change)
Money multiplier (% change)
***
***
**
***
***
***
*
***
Main Findings on Monetary Performance




While inflation objectives are generally met,
Money growth is higher than projected
With significantly higher reserves – in part
due to the lack of absorption of aid inflows
Which is partly offset by lower domestic
credit – in part due to sterilization using
domestic instruments
The Post-Stabilization LICs
Country list:
Albania, Azerbaijan, Bangladesh, Benin,
Ethiopia, Guyana, Honduras, Kyrgyz
Republic, Madagascar, Mongolia,
Mozambique, Rwanda, Senegal, Tanzania,
and Uganda.
Concluding Thoughts



“Spend and absorb” is the only sensible response to
aid in the long run.
– Some real appreciation may be necessary to enable
reallocation.
– When maintaining a peg, some inflation may be
necessary as a relative price adjustment (and
reallocation of resources).
– In the short run, other options may be considered
(e.g., building reserves)—as in Ghana, Ethiopia
“Spend and not absorb” is problematic option because
it leads to deficit financing. If aid is to increase
reserves, it can’t also finance spending. Yet this is a
common response.
Ensure coordination between fiscal and monetary
policies.