Uploaded by Moises Ruiz de

Lecture 5 24102022 (1)

advertisement
Global Economic Environment
and Perspectives:
the economics of money and banking.
Liliya Repa, PhD
Lecture #5
24 October, 2022
1
Social
Analysis
Economic
Analysis
Financial
Analysis
Equity
Analysis
Producing credible
evidence to answer policy
or programmatic
questions to determine
whether an intervention
is worthwhile
Impact
Analysis
2
Levels of Analysis
Activity
Macro-level
Country Level
country
political
economy
assessment
Thematic
analysis: (e.g.,
analysis of
Sector/Theme
natural resource
management)
PE Analysis
PE analysis of
of core
other sectors
governance
(e.g., water,
issues &
transport)
reform
PE to inform
Project a specific
project or
component
PE focused
on a single
policy
decision
Drivers of
decision
making at
the national
influence
sector level
dynamics
3
Macro economic objectives
1. External Balance
• Balance of payment
• Current account
• International reserves
• Exchange rate stability
2. Internal Balance
High employment
Low inflation
Sustainable growth
4
Economic policy instruments
Fiscal policy
• Changes in Government expenditure
• Changes in tax system
• Tax-financed
• Debt-financed
Monetary policy
• Changes in domestic monetary policy
• Changes in the stock of domestic bonds
• Changes in the stock of foreign reserves
Exchange rate
• Flexible exchange rate
• Fixed exchange rate
• Managed float, crawling peg
5
Monetary Policy: Limitations
1. The short-term nominal interest rate (policy rate) cannot go
below zero (“zero lower bound”)
• when the economy is in a slump, a nominal interest rate of
zero may not be low enough to stabilize the economy
• Quantitative easing = Central bank purchases of financial
assets aimed at increasing investment by reducing yields.
2. A country without its own currency does not have its own
monetary policy
• E.g. countries of the eurozone
Capacity Constraints
Another reason for the inflationunemployment trade-off are capacity
constraints.
Firms respond to rising capacity
utilization by increasing investment. In
the short run, firms are capacity
constrained (unable to meet excess
demand for output) so raise prices.
Wage-price spiral when other firms
respond in the same way.
Price rigidities
• Price stickiness (or sticky prices) is the resistance of
market price(s) to change quickly despite changes in the broad
economy that suggest a different price is optimal.
• ... Price stickiness can also be referred to as "nominal rigidity"
and is related to wage stickiness.
8
Financial frictions
• is the difference between the return businesses earn from
capital—plant and equipment—and the market cost of capital.
9
Long-run Philipps curve
• The Phillips curve depicts the relationship between inflation
and unemployment rates. The long-run Phillips curve is a
vertical line that illustrates that there is no permanent trade-off
between inflation and unemployment in the long run. .
10
INTEREST RATE POLICY UNDER THE TAYLOR RULE
11
FROM THE SHORT RATE TO THE LONG RATE
• The problem: the central bank may control the short-term interest rate, but aggregate demand mainly depends
on the long-term interest rate.
• Assumption: Short-term and long-term bonds are perfect substitutes
•
This implies the Arbitrage condition
(1  itl )n  (1  it )  (1  ite1 )  (1  ite 2 )  ........  (1  ite n 1 )
• Taking logs on both sides and using the approximation ln(1+i)  i, we get
•
The expectations theory of the term structure of interest rates
1
i  (it  ite1  ite 2  ......  ite n 1 )
n
l
t
Implication: The current long rate is a simple average of the current short
rate and the expected future short rates.
12
itl  it
iff
ite j  it
for all j  1, 2,..., n  1
13
DERIVING THE AGGREGATE DEMAND CURVDERIVING THE AGGREGATE DEMAND CURVE
The ex post real interest rate
1 i
1 r 
1   1
a
 r a  i   1
The ex post real interest rate
Investment and consumption are governed by
(27)
Investment
and
consumption
are governed
by
The ex
ante
real interest
rate
rate
1  i The ex ante real interest
e
1 r 
 r  i   1
e
1   1
(28)
We assume We assume
Static expectations
Static expectations
Equations (28) and (29) imply
 e1  
(29)
r  i 
(30)
DERIVING THE AGGREGATE DEMAND CURVDERIVING THE AGGREGATE DEMAND CURVE
y  y  1 ( g  g )   2 (r  r )  v,
1  0,
i  r    h(   *)  b( y  y ),
h  0,
2  0
b0
(11)
(21)
we get
The aggregate demand curve
r r
y  y  1 ( g  g )   2 [h(   *)  b( y  y )]  v 
y  y   ( *  )  z
2h

0
1   2b
v  1 ( g  g )
z
1   2b
(32)
(33)
PROPERTIES OF THE AGGREGATE DEMAND
CURVE
 The AD curve has a negative slope: higher inflation induces the
central bank to raise the interest rate, causing aggregate demand to fall
 The AD curve is flatter, the more weight the central bank attaches to
stable inflation compared to output stability
The AD curve shifts upwards in case of more optimistic growth
expectations in the private sector or in case of a more expansionary fiscal
policy
 The AD curve shifts downwards if the central bank reduces its inflation
target
Unconventional monetary policies
• Quantitative Easing
• What is it and why is it important?
• Why are banks holding so many reserves?
• Assessment: useful? Inflationary?
• Exit
• Spillover effects on emerging markets
• Forward guidance
• Negative interest rates
17
Quantitative easing
• Quantitative easing (QE), also known as large-scale asset purchases, is
a monetary policy whereby a central bank buys predetermined amounts
of government bonds or other financial assets in order to
inject liquidity directly into the economy. An unconventional form of
monetary policy, it is usually used when inflation is very low or negative,
and standard expansionary monetary policy has become ineffective. A
central bank implements quantitative easing by buying specified amounts
of financial assets from commercial banks and other financial institutions,
thus raising the prices of those financial assets and lowering their yield,
while simultaneously increasing the money supply. This differs from the
more usual policy of buying or selling short-term government bonds to
keep interbank interest rates at a specified target value.
Central Banking
Assets
Net foreign assets
Net domestic assets
Net domestic credit
- Net claims on
government
- Claims on MDBs
- Claims on other
sectors
Other items, net
Liabilities
Reserve money
Currency in circulation
- Held in banks
- Held outside banks
Others liabilities
19
Control of the Monetary Base
MB = C + R
Open Market Purchase from bank
The Banking System
Assets
Liabilities
The Bank
Assets
Liabilities
Securities – $100
Reserves + $100
Open Market Purchase from Public
Public
Assets
Liabilities
Securities + $100
Reserves + $100
The Bank
Assets
Liabilities
Securities – $100
Deposits + $100
Banking System
Assets
Securities + $100
Reserves + $100
Liabilities
Reserves
+ $100
Cheqable Deposits
+ $100
Result: R  $100, MB  $100
If Person Cashes Cheque
Public
Assets
Liabilities
Securities – $100
Currency + $100
The Bank
Assets
Securities + $100
Liabilities
Currency + $100
Result: R unchanged, MB  $100
Effect on MB certain, on R uncertain
The effect of an open market purchase on R depends on whether the
seller of the bonds keeps the proceeds from the sale in deposits or in
currency.
The effect of an open market purchase on MB, however, is always the
same whether the seller of the bonds keeps the proceeds from the sale in
deposits or in currency.
Open Market Sale of Bonds
MB = C + R
Open Market Sale to a bank
The Banking System
Assets
Liabilities
The Bank
Assets
Liabilities
Securities + $100
Reserves - $100
Open Market Sale to the Public
Public
Assets
Liabilities
Securities - $100
Reserves - $100
The Bank
Assets
Liabilities
Securities + $100
Deposits - $100
Banking System
Assets
Securities - $100
Reserves - $100
Liabilities
Reserves
- $100
Cheqable Deposits
- $100
Result: R  $100, MB  $100
Open Market Purchase in the Foreign Exchange (FX)
Market
MB = C + R
Open Market Purchase of foreign exchange from a bank
The Banking System
Central bank
Assets
Liabilities
Assets
Liabilities
FX – $100
FX + $100
Reserves + $100
Open Market Purchase of foreign exchange from the Public
Public
Central bank
Assets
Liabilities
Assets
Reserves + $100
FX – $100
Deposits + $100
Banking System
Assets
Reserves + $100
Liabilities
Reserves
+ $100
Chequable Deposits
+ $100
FX + $100
Result: R  $100, MB  $100
Liabilities
If Person Cashes Cheque
Public
Assets
FX – $100
Currency + $100
Liabilities
The Bank
Assets
FX + $100
Liabilities
Currency + $100
Result: R unchanged, MB  $100
Effect on MB certain, on R uncertain
Again, the effect of an open market purchase on R depends on whether
the seller of FX keeps the proceeds from the sale in deposits or in
currency.
The effect of an open market purchase of FX on MB, however, is always
the same whether the seller of the FX keeps the proceeds from the sale in
deposits or in currency.
Open Market Sale in the FX Market
MB = C + R
Open Market Sale of FX to a bank
The Banking System
Assets
Liabilities
The Bank
Assets
Liabilities
FX + $100
Reserves -$100
Open Market Sale of FX to the Public
Public
Assets
Liabilities
FX - $100
Reserves - $100
The Bank
Assets
Liabilities
FX + $100
Deposits - $100
Banking System
Assets
FX - $100
Reserves - $100
Liabilities
Reserves
- $100
Cheqable Deposits
- $100
Result: R  $100, MB  $100
Shifts from Deposits into Currency
Even if the Central Bank does not conduct open market operations, a shift
from deposits into currency will affect R. However, such a shift will have
no effect on MB.
Shifts From Deposits into Currency:
Public
Assets
Liabilities
The Bank
Assets
Deposits – $100
Currency + $100
Banking System
Assets
Liabilities
Reserves – $100 Deposits – $100
Result: R  $100, MB unchanged
Liabilities
Currency + $100
Reserves – $100
Advances
Banking System
The Bank
Assets
Liabilities
Assets
Liabilities
Reserves
Advances
Advances
Reserves
+ $100
+ $100
+ $100
+ $100
Result: R  $100, MB  $100
Conclusion: Bank has better ability to control MB than R
Deposit Creation: Single Bank
Consider a $100 open market purchase from First Bank
First Bank
Liabilities
Assets
Securities
Reserves
– $100
+ $100
First Bank
Liabilities
Assets
Securities
Reserves
Loans
– $100
+ $100
+ $100
Deposits
+ $100
…
A bank cannot safely make loans for an amount greater than the
excess reserves it has before it makes the loan.
The final T-account of the First Bank (after the reserves have been
withdrawn) is:
First Bank
Assets
Securities
Loans
Liabilities
– $100
+ $100
Deposit Creation: Banking System (r = 10%)
Assets
Reserves
+ $100
Assets
Reserves
Loans
+ $10
+ $90
Assets
Reserves
+ $90
Assets
Reserves
Loans
+$9
+ $81
Bank A
Liabilities
Deposits
Bank A
Liabilities
Deposits
Bank B
Liabilities
Deposits
Bank B
Liabilities
Deposits
+ $100
+ $100
+ $90
+ $90
Simple Deposit Multiplier
Simple Deposit Multiplier
D =
1
 R
r
Deriving the formula
R = RR = r  D
D=
D =
1
r
1
r
R
 R
Multiple Deposit Contraction
The multiple deposit creation process should also work in reverse.
When the Central Bank withdraws reserves from the banking system
, there should be a multiple contraction of deposits.
In fact, the contraction in deposits will be
D = (1/ r )  R
Example:
If R = -100 and (1/ r ) = 10 because r =.10, then
D = -1000.
Multiple Deposit Contraction: The Banking
System
Banking System
Assets
Liabilities
Securities + $100
Deposits
Reserves - $100
Loans
- $1000
- $1000
The Bank is not the only player whose behavior influences D
and M.
D and M depend on:
1. the public’s decisions regarding how much C to hold
2. the banks’ decisions regarding the amount of R they wish
to hold, and
3. borrowers’ decisions on how much to borrow from banks.
35
• According to several authors
the policy had a significant
impact reducing the risk
premiums
•
• TALF and asset purchases in
the U.S. led to the return of
liquidity in the securitized credit
markets
36
Will inflation increase?
Under traditional operating framework the central bank had to remove all of the
excess reserves to stop the process of “money creation” (loans and deposits) by the
banks
Now reserves are not necessarily inflationary because paying interest on reserves
breaks the link between quantity of reserves and bank’s willingness to lend to firms
and households
Central bank paying interest on reserves is equivalent to issuing own securities
(debt) to sterilize the increase in liabilities
Some of the original programs were already allowed to expire like facility to lend
directly to financial institutions
37
The impact of unconventional measures on other countries depends on
• Cyclical position of the recipient
• Market imperfections
• Policy and supervision
• Ultimately, on the strengths and vulnerabilities of each economy
• Sound macroeconomic policies are essential to avoid potential negative effects of
spillovers, monetary and fiscal policy mix, exchange rate flexibility, build up of reserves,
proper financial sector supervision
• And as spillovers are more prevalent and larger if they are the result of signal surprises
(Sahay and others, 2015)
38
UMP Forward guidance
• With these extraordinary measures communication of future policy became very
important to guide inflation expectations
• Not a new element of monetary policy, even at zero lower bound
• Many inflation forecast targeters use forward guidance regularly
• Forward guidance at the ZLB was first adopted by the Bank of Japan in the context
of its zero interest rate policy in 1999
• Filardo and Hofmann (BIS QR Mar 2014)
— Provide additional stimulus at the ZLB when central banks communicate that
policy rates will remain lower for longer could reduce long term rates (mortgages)
— Reduce uncertainty, thereby lowering interest rate volatility and through this
channel possibly also risk premium
39
UMP Forward guidance: delayed lift off
In the presence of zero lower bound and high uncertainty optimal monetary policy is
• A delayed lift off of policy rates and
• A modest, but planned, overshooting of inflation (see Aichi, Laxton and others, 2015, “Avoiding
Dark Corners”)
It requires publication of
• A baseline forecast and
• A description of uncertainties around that outlook
• Combined with enhanced communications (of endogenous policy rate path and risks)
In a situation of persistent output and inflation this building of “buffer stock” is clearly
optimal (Evans and others, 2015)
• Question: possible accumulation of financial vulnerabilities (“leaning against the wind” policy
debate)?
40
QE in EU and Japan: Negative Policy Interest Rates
• Central banks (the Bank of Japan and the European CBs), have also implemented
unconventional monetary policy
• They designed programs to buy assets to expand liquidity and keep short and
long-run interest rates at historical low levels
• They went even further than the FED and implemented negative interest rates
• Banks are charged a fee on reserves above a threshold instead of receiving
interest on reserves they hold with the central bank
• To avoid the fee, banks would shift to other short-term assets, which drives down
the yields on those assets as well
41
QE in EU and Japan:
Negative Policy Interest Rates
• Negative rates for reserves and possibly other assets create a profit squeeze for banks
• Beyond a certain limit commercial banks would move to holding cash instead of paying
the fee. Expected tipping point when banks would move into cash
• 75 basis points (IMF staff estimates)
• Cost of using cash as a store of value or for large transactions (vault capacity, cash
transport, payment systems set up)
• Some costs are one-off and length of time of negative rates matters
• Costs are country specific (such as highest denomination of bank notes)
• Political economy and social limits
• Feeling of being “taxed” or “betrayed”
42
The Monetary Policy Framework
Closely related with two other policies:
1. The conditions that specify how a country’s nominal
exchange rate is determined (fixed, floating, managed,
etc)—Exchange Rate Regime.
2. Degree of openness of the capital account—
Capital Account Policies
43
Exchange Rate: What is it?
Price of one national money in terms of another
Example: E $/Pound = 1.60
E yen/$ = 120
Need convention: Direct Quote
E = home currency price of foreign currency
Depreciation: Rise in E
Appreciation: Fall in E
Three Commonly Quoted Rates:
Spot: Et delivery 2 days
Forward: Ft, N delivery in N days
Swap: Spot Purchase (sale) -Forward Sale (purchase) Given
amount of foreign currency
44
Exchange rate
Exchange rate = number of units of home currency that can
be exchanged for one unit of foreign currency.
Interest rates affect demand for home currency in the
foreign exchange market, so affects the exchange rate
(appreciation/depreciation).
The exchange rate affects relative demand for homeproduced goods, so affects net exports.
Therefore, interest rates affect aggregate demand through
the market for financial assets.

The Purchasing Power Parity Theory. While it can be expressed differently, the most common
expression links the changes in exchange rates to those in relative price indices in two
countries:
Rate of change of exchange rate = Difference in inflation rates

The International Fisher Effect (IFE). This holds that an interest rate differential will exist only
if the exchange rate is expected to change in such a way that the advantage of the higher interest
rate is offset by the loss on the foreign exchange transactions. Practically speaking, the IFE
implies that while an investor in a low-interest country can convert funds into the currency of a
high-interest country and earn a higher rate, the gain (the interest rate differential) will be offset by
the expected loss due to foreign exchange rate changes. The relationship is stated as:
Expected rate of change of the exchange rate = Interest rate differential

The Unbiased Forward Rate Theory. This holds that the forward exchange rate is the best and
unbiased estimate of the expected future spot exchange rate:
Expected exchange rate = Forward exchange rate
46
The exchange rate risks these factors create can be arranged into three primary categories:

Economic exposure. Due to changes in rates, operating costs will rise and make a
product uncompetitive in the world market, thus eroding profitability. There’s little
that can be done about economic risk; it’s simply a routine business risk that every
enterprise must endure.

Translation exposure. The impact of currency exchange rates will reduce a
company’s earnings and weaken its balance sheet. In turn, the denominations of
assets and liabilities are important, although many experts contend that currency
fluctuations have no significant impact on real assets.

Transaction exposure. Caused by an unfavorable move in a specific currency
between the time when a contract is agreed and the time it is completed, or
between the time when lending or borrowing is initiated and the time the funds
are repaid. This is the most common problem that confronts most companies.
Requiring payment in advance is rarely practical, and impossible, of course, for
borrowing and lending.
47
Euro (EUR) to U.S. dollar (USD) exchange rate from January 1999 to
October 19, 2022
Exchange rate as transmission mechanism
Exchange rate policy principles
• Since April 12, 2000 the zloty exchange rate has been a floating exchange
rate that is not subject to any restrictions. The central bank does not aim to
set predetermined zloty exchange rates against other currencies. It reserves,
however, the right to intervene if it deems this necessary in order to achieve
the inflation target.
• On its accession to the European Union, Poland undertook to join the euro
zone. Thus in the future the zloty will be replaced with the common
European currency, and monetary policy will be shaped by the European
Central Bank.
• Meeting the exchange rate stability criterion is one of the conditions of
joining the euro zone. Therefore before the adoption of the euro, the zloty
exchange rate against the euro remains fixed for at least two years within
the (Exchange Rate Mechanism II). This means that during this period
Narodowy Bank Polski will maintain the market zloty exchange rate against
the euro within the permissible range, with regard to the set central parity.
50
Determination of exchange rate:
LR
Equilibrium characterized by CA imbalances such as to keep NIIP/GDP constant. Level of equilibrium NIIP/GDP determined by factors
considered intertemporal theory of c/a. Steady-state is independent of price level.





SR



Shocks are monetary
PPP prevails between equilibrium
the real shocks are of second order importance
in LR something close to PPP holds in LR, but LRER, NFA, TOT, G/Y, productivity
Leads to theory of debt crises.
UIP
Overshooting
Portfolio models
• Exchange rates response ( changes in fundamentals (MS, income level, interest rates, expected inflation rates, TOT, productivity)
• Exchange rate changes are disconnected from fundamentals, but at level co integrated with fundamental value.
Foreign interest rates raise
Domestic collateral fall
high collateral
Medium collateral
Low collateral
SR
expansion
contractionary
expansion
contractionary
expansion
LR
expansion
contractionary
mixed
expansion
expansion
51
Central Bank and Exchange Rates
• A central bank can intervene in exchange markets in
two ways:
It can raise or lower interest rates to make the currency
stronger or weaker.
Or it can directly purchase or sell its currency in foreign
exchange markets.
52
Exchange Rate Intervention, Sell $
1. Sell $, buy F: MB , Ms 
2. Ms , P , Eet+1 , expected appreciation of
F , RF shifts right in
3. Ms , iD , RD shifts left, go to point 2 and Et

4. In long run, iD returns to old level, RD shifts
back, go to point 3: Exchange rate
overshooting
Foreign exchange interventions
• NBP may carry out interventions in the FX market
54
The Gold Standard
Currency convertible into gold at fixed value
Example of how it worked:
Canada: $20 converted into 1 ounce
U.K.: £4 converted into 1 ounce
Par value of £1 = $5.00
If £  to $5.25, importer of £100 of tweed has two alternatives:
1. Pay $525
2. Buy $500 gold (500/20 = 25 ounces), ship to U.K., convert into
£100 (= 25 £4) and buy tweed
The Gold Standard
If shipping cheap, do alternative 2
1. Gold flows to U.K.
2. MB  in U.K, MB  in Canada
3. Price level  U.K.,  Canada
4. £ depreciates back to par
Two Problems:
s
1.
Country on gold standard loses control of M
2.
World inflation determined by gold production
Fixed Exchange Rate Systems
Bretton Woods
1.Fixed exchange rates
2.Other central banks keep exchange rates fixed to $: $ is reserve currency
3.$ convertible into gold for central banks only ($35 per ounce)
4.International Monetary Fund (IMF) sets rules and provides loans to deficit countries
5.World Bank makes loans to developing countries
European Monetary System
1.Value of currency not allowed outside “snake”
2.New currency unit: ECU
3.Exchange Rate Mechanism (ERM)
Key weakness of fixed rate system
Asymmetry: pressure on deficit countries losing international reserves to  M, but no
pressure on surplus countries to  M
Intervention in a Fixed Exchange Rate System
F
Since Eet+1 = Epar with fixed exchange rate, R doesn’t shift
Overvalued exchange rate (panel a)
1.
Central bank sells international reserves to buy domestic
currency
2.
MB , M , i , R to right to get to point 2
3.
If don’t do this, have to devalue
s
D
D
Undervalued exchange rate (panel b)
1.
Central bank sells domestic currency and buys international reserves
2.
MB , M , i , R to left to get to point 2
3.
If don’t do this, have to revalue
s
D
D
Role of a Nominal Anchor
Ties Down  Expectations
Helps Avoid Time-Consistency Problem
1. Arises from pursuit of short-term goals which lead to bad long-term
outcomes
2. Time-consistency resides more in political process
3. Nominal anchor limits political pressure for time-consistency
Exchange-Rate Targeting
Disadvantages
1. Loss of independent monetary policy
Problems after German reunification: UK, French monetary policy
too tight
2. Open to speculative attacks
Europe, Sept. 1992; Mexico: 1994; Asia: 1997
3. Successful speculative attack disastrous for emerging market
countries because it leads to financial crisis
4. Weakened accountability: lose exchange-rate signal
Currency Boards vs. Dollarization
Currency Boards
1. Domestic currency exchanged at fixed rate for foreign
currency automatically
2. Fixed exchange rate with very strong commitment
mechanism and no discretion
3. Usual disadvantages of fixed exchange rate
4. Still subject to speculative attack
5. Lose ability to have lender of last resort
Dollarization
1. Even stronger commitment mechanism
2. No possibility of speculative attack
3. Usual disadvantages of fixed exchange rtae
4. Lose seignorage
Weighted average monthly exchange rate of
Eur, USD and CHF
62
• Understanding the mechanism of external shocks transmission to
domestic prices is crucial for ensuring stable economic development
and low inflationary outlook of any open economy
• Exchange Rate Pass-Through (ERPT) – is the percentage change, in
local currency, of import prices resulting from a one percent change
in the exchange rate between the exporting and importing countries
– Goldberg & Knetter (1997)
• Also commonly used to express the effect of exchange rate
movements to other price indices (producer and consumer prices)
• Being an open economy, country is sensitive to external shocks
Direct and indirect channels of exchange rate pass-through mechanism.
Evolution of exchange rate pass through:
• Empirical literature finds that ERPT to domestic prices is far from complete even in the long run (see Menon (1995))
Two stands of theoretical literature:
Pricing-to-market and imperfect competition
• Foreign exporting firm under the imperfect competition conditions has a pricing power on the importing country’s market and tend to adjust their
mark-ups in response to exchange rate fluctuations (see Dornbusch (1987) and Fischer (1989))
• Mark-up responsiveness will depend mainly on the market share of domestic producers relative to foreign producers
Currency pricing strategy
• The degree of ERPT depends on the pricing strategy of firms: producer (PCP) and local currency pricing (LCP) (see Betts & Devereux (1996))
• Under the PCP, when the price is set in the currency of exporter, FX movements are fully reflected in the price of imported product expressed in the
local currency, resulting in complete ERPT
• LCP implies that prices are pre-set in domestic currency and ERPT is zero
• The aggregate pass-through depends on the combination of firms with different pricing strategies.
• the extent to which domestic prices respond to exchange rate fluctuations varies among countries.
the underlying determinants of ERPT.
Traditionally well explained by a set of macroeconomic factors:
• country’s size and openness (see Goldfajn & Werlang (2000), McCarthy (2000)),
• import composition (see Campa & Goldberg (2005)),
• inflation environment and monetary policy (see Taylor (2000), Bailliu & Fujii (2004), Choudhri & Hakura (2001, 2006)),
• exchange rate regime (as in Beirne (2009)) • others
• cross-country spillovers
NOTATION
PTR: Home currency Price of a basket of traded goods purchased by home consumers
P*TR*: Foreign currency Price of a basket of traded goods purchased by foreign consumers
PT: Home currency price of home produced tradable
PT*: Home currency price of foreign produced tradable
P*T: Foreign currency price of home produced tradable
P*T*: Foreign currency price of foreign produced tradable
PN: Home currency price of home produced non-tradable
P*N*: Foreign currency price of foreign produced non tradable
TOT: Terms of trade, denoted by Q, relative price of imports in terms of exports
δ: The real CPI exchange rate
η: Relative price of non traded goods in terms of consumption basket traded goods
ρ: Relative price of non traded goods in terms of domestic exportables
LOP: Law of One Price
67
67
Notes on Exchange Rate Building Blocks
Exchange Rates, Price Levels, and Relative Prices (Cobb Douglass Benchmark)
Baskets of Traded Goods
PTR  [ PT A PT *1 A ]  PT [TOT ]1 A
P * TR*  [ P * T A P * T *1 A ]  P * T *[TOT *] A
*
where
*
*
TOT = PT * /PT
TOT* = P*T / P*T *
If LOP
PT* = EP*T *
P*T = PT / E
TOT = 1 / TOT *
P*TR* = P*T *[TOT]- A
*
With same baskets and LOP
PTR= EP*TR*
Consumer Price Indices
CPI = PTRBPN1-B = PTR[PN / PTR]1-B
CPI * = P*TR*B P* N *1-B = P*TR*[P* N * /PTR*]1-B
*
Absolute PPP
Relative PPP
*
ECPI * /CPI =1
ECPI * /CPI = d
*
With Identical Weights, Identical Baskets, and LOP
ECPI * /CPI = [P* N * /P*TR*]1-B /[PN / PTR]1-B
E = [CPI / CPI *][P* N * /P*TR*]1-B /[PN / PTR]1-B
Departure from Absolute PPP
E = [CPI / CPI *][h * /h ]1-B
d = [h * /h]1-B
Where are Terms of Trade?
P* N * /P*TR* = [P* N * /P*T*]TOT A
PN / PTR= [PN / PT]TOT A-1
Thus
E = [CPI / CPI *][P* N * /P*T*]1-B[PN / PT]1-B[TOT]1-B
E = [CPI / CPI *][ r * / r ]1-B Q1-B
Departure from Absolute PPP
d = [ r * / r ]1-B Q1-B
Home Bias with all Goods Traded and LOP
CPI = PTR= PR[TOT]1-A
CPI * = P*TR* = P*T *[TOT]- A*
Departure from Absolute PPP
If A>A*, we have Home Bias
E = [CPI / CPI *]TOT A-A*
d = QA-A*
Monetary Approach to Exchange Rate Determination
Quantity Equation
MV  PY  PGDPYGDP
Departure from PPP
Combining
*
*
M *V *  P * Y *  PGDP
YGDP
EP * / P  
E  [ M / M *][Y * / Y ][V / V *]
Note that the exchange rate depends on monetary and real factors. Here P is CPI and Y is nominal GDP
deflated by the exact CPI. Quantity Equation with Elastic Velocity
Mv[1+ i]l = PY
M * v*[1+ i*]l = P*Y *
We now get
E  [M / M *][Y / Y *][v / v*][(1  i) /(1  i*)] 
It appears we have added another fundamental. But by UIP
[(1  i) /(1  i*)]  E e / E
Combining
E  [M / M *][Y * / Y ][v / v*][ E e / E ] 
This is the key equation of the monetary approach to exchange rate determination. Taking logs of both
sides we obtain
1
e
et  (1   ) [( mt  mt *)  ( yt *  yt )  (vt  vt *)   t ]   /(1   )e
t 1
Thus, the log of the exchange rate is a weighted average of the log of a
composite fundamental and the log of the expected future exchange rate.
This equation can be solved forward to obtain
et = (1+ l )
-1
where
¥
å(l / (1+ l )) z
i e
t+i
i=0
zt  (mt  mt *)  ( yt *  yt )  (vt  vt *)   t
is the log of the composite fundamental. Thus, according to the monetary approach, the exchange rate is an asset
price that is discounted present value of current and expected future fundamentals.
The fundamentals are home and foreign money supplies and money demands, home and foreign outputs, and the
equilibrium deviation from absolute PPP .
Thus even according to the “monetary” approach, real factors such as productivity and demand (for nontraded
goods or traded goods that are not perfect substitutes) are expected to alter equilibrium nominal exchange rates.
Implications
If the fundamentals are constant, the exchange rate is constant and equal to the (composite) fundamental.
If the composite fundamental is I(1), it and the exchange rate and cointegrated.
If the composite fundamental is I(1) and its growth rate is¥ persistent, the effect of a shock to the fundamental has
a magnified effect on the exchange rate.
example,
e -For
z=
(1+ l )-1if (l / (1+ l ))i Dze
t
t
¥
t+i
i=0
Then
Dzt+1 = rDzt + et
¶et / ¶et = (1+ l ) / (1+ l - rl )
We can write the monetary approach as a two equation system
E  [ M / M *][Y * / Y ][v / v*][(1  i) /(1  i*)]
[(1  i) /(1  i*)]  E e / E
Holding constant the composite fundamental Z = [M/M*][Y*/Y][v/v*]
The exchange rate is an increasing function of (one plus) the interest differential. (One plus) the interest
differential, holding constant the expected future exchange rate, is a decreasing function of the current exchange
rate.
Suppose that there are two periods, present and future and that the expected future exchange rate is Ee = Zf. If Z
in the present is different from Zf, it will influence the present exchange rate but not the future.
A temporary, present rise in M (that leaves Zf unchanged) shifts up the EE curve leading to a depreciation of the
present exchange rate and a fall in the home interest rate relative to the foreign interest rate. Since UIP and
relative PPP hold, expected inflation at home must fall. Why, because the home price level rises in the present
leading to expected deflation at home.
A temporary present rise in Y (that leaves Zf unchanged) shifts down the EE curve (holding δ constant and equal
to one) leading to an appreciation of the present exchange rate and a rise in the home interest rate relative to the
foreign interest rate. Since UIP and the home price level falls in the present leading to expected inflation at
home.
A temporary, present rise in δ (that leaves Zf unchanged) shifts up the EE curve leading to a depreciation of the
present exchange rate and a fall in the home interest rate relative to the foreign interest rate. The home real
interest rate also falls below the foreign real interest rate.
A permanent, future rise in M (that leaves Z unchanged) shifts up the UIP curve leading to a depreciation of the
present exchange rate and a rise in the home interest rate relative to the foreign interest rate. Since UIP and
relative PPP hold, expected inflation at home must rise because the home price level in future rises.
A permanent future rise in Y (that leaves Z unchanged) shifts down the UIP curve (holding δ constant and equal
to one) leading to an appreciation of the present exchange rate and a fall in the home interest rate relative to the
foreign interest rate. Since UIP and relative PPP hold, expected inflation at home must fall because the home
price level falls in the future leading to expected deflation at home.
A permanent, future rise in δ (that leaves Z unchanged) shifts up the UIP curve leading to a depreciation of the
present exchange rate and a rise in the home interest rate relative to the foreign interest rate. The home real
interest rate must rise above the foreign real interest rate.
Foreign exchange swaps
• Foreign exchange swaps - foreign exchange swap is a transaction in
which central bank purchases (or sells) national currency against
foreign currency in the spot market and, at the same time, resells it
(or repurchases) under a forward contract at a specified date.
74
75
Choosing a regime
Different exchange rate regimes can help achieve different objectives:
• Flexible exchange rate regimes allow policymakers to use monetary policy to achieve
domestic price and macro stability objectives. Greater policy independence (in principle).
• Facilitate adjustment to external shocks, through the expenditure switching channel
(though it may depend on the type of shock).
• A fixed exchange rate can serve as a nominal anchor, obviating the need for an
independent and credible monetary policy (which may be difficult to achieve).
• A corollary is that fixed exchange rates are more susceptible to crises and speculative
attacks, as imbalances can accumulate over time and pegs can become unsustainable.
• Fixed exchange rates can promote greater trade and financial integration, by reducing
transaction costs and the uncertainty related to exchange rate volatility.
• The choice of regime reflects policymakers’ assessment of the relative importance of
these objectives for their country.
76
Monetary Policy Frameworks and Exchange
Regimes
𝑒
• In the UIP replace model consistent expectations 𝐸𝑡 𝑠𝑡+1 by 𝑠𝑡+1
…
st  ste1  (it*  it  premt ) / 4   ts
• …we allow for some “backward-lookingness” in the expected NER
ste1  (1  e1 ) Et st 1  e1 ( st 1 
Model-consistent
expected NER
Past NER
2
st )
4
The “long-run”
change in NER
• We set 0 < 𝑒1 <1 to account for NER persistence
• The long-run change in the NER is consistent with the long-run inflation differential and
the change in equilibrium RER
s   T   *  z
Domestic long-run
inflation rate –
inflation target
Foreign long-run
inflation rate –
inflation target
The “long-run”
change in RER
77
Triangular Arbitrage and the Vehicle Currency
Exercise .
The CAD/EUR exchange rate (a small illiquid market) is pinned down by the CAD/USD and USD/EUR
markets which are much more liquid by TRIANGULAR ARBITRAGE.
CAD/EUR = (USD/EUR)(CAD/USD)
1.5051 = 1.0835 times 1.3892
Suppose this did not hold with CAD/EUR at 1.49
USD/EURO > (CAD/EUR)/(CAD/USD)
Then Take 100 USD and buy 138.92 CAD. Take these 138.92 CAD and buy 138.92/1.49 EUR. Take these
Eur and sell them for (138.92/1.49)1.0835 dollars.
You end up (1 second later) with $101.02. That’s $1.02 bps of profit a second with no risk. $838 of profit per
minute (and 83.8 % return!) a minute. $2,221,1858 of profit a day a day. At no risk. So why stop at $100?
78
How bout a billion? All day. Thus triangular arbitrage must hold. And it does.
The Real Exchange Rate
Price of one national output in terms of another:
Q = US Goods/German Goods
Real Depreciation: Rise in Q
Real Appreciation: Fall in Q
Key equation linking real and nominal exchange rate
Q = EP*/P
Nominal Exchange Rate: E
Foreign Price level: P*
US Price Level: P
When E rises (falls) by more than P/P*, there is a real
depreciation (appreciation)
79
Some Key Facts
Over long periods of time, exchange rate changes reflect inflation differentials
Floating exchange rates are much more volatile than national price levels
Exchange Rates wander away from national price levels for long periods of time
80
https://www.portfoliovisualizer.com/
monte-carlo-simulation
https://www.mataf.net/en/forex/tools
/martingale
Purchasing Power Parity: A Theory Linking Exchange Rates and Price
Levels
According to theory of absolute PPP, we should see (on average)
E = P/P*.
Price of goods equal (on average) when expressed in common currency
P =EP*.
Thus Absolute PPP is theory that in long run real exchange rate Q = 1.
If PPP held exactly, we would expect E and P*/P to coincide.
83
Relative PPP
According to theory of relative PPP, we should see (on average)
Et = (Pt/P*t)Q
The real exchange rate Q is constant but not necessarily equal to 1
Et P*t/Pt = Q
In log differences, relative PPP (ex post)
∆Et,1/Et = ∆Pt,1 - ∆P*t,1 = πt,1 – π*t,1
If this holds in expectation , relative PPP (ex ante)
∆E et,1/Et = πet,1 – π*et,1
84
How do we Interpret departures from absolute PPP?
• Countries produce different goods (Fords and Toyotas)
Can be due to changes in the equilibrium relative price of these goods
But can also result from volatile exchange rates and 'sticky' price levels
With no change in (long run) equilibrium relative price
85
Interest Rates and Bond Yields
Key Fact: Bond yields and interest rates Are Not Equalized Internationally
A Free Lunch? Arbitrage Profit?
When an investor buys a foreign bond, his total return depends on the exchange rate change
Even if foreign currency proceeds known, home currency value is not
Expected returns can be equalized even if bond yields and interest rates differ
The fact that ex post returns differ may be due to exchange rate surprises
86
Uncovered Interest Rate Parity
Consider the dollar return to buying a US bond that matures in 1 year.
1 $ today ~> (I + R t, $) $ in 1 year
Consider the dollar return to buying a Sterling bond that matures in 1 year.
First, today buy sterling
1$ today = (I / Et) GBP today
Second, Invest today in sterling deposit
(1/ Et) GBP today ~> (1 + Rt.GBP) / Et GBP in 1 year
Third, wait for a year, collect principal and interest, and then sell proceeds
for USD in spot market at spot rate Et+1
(I + Rt, GBP) / Et GBP in year = (1 + Rt, GBP)(Et+ I / Et) USD in 1 year
87
Key Insight:
As of to day's date t, R t, $ R t, GBP and Et are known, but Et+1 is not.
The investors know his pound return to investing in UK
but does not know his dollar return. His expected dollar return is
(1 + Rt, GBP)(E et+ 1 / Et)
where E et+ 1 is the expected exchange rate in 1 year.
Under what condition will the expected dollar return to investing in pounds
equal the known dollar return to investing in US?
Answer: When
(E et+ I / Et) = (1 + Rt, $)/ (1 + Rt, GBP)
88
Uncovered Interest Parity
Is the hypothesis that expected returns to international investing
are equalized regardless of the currency of denomination of the
investment.
If UIP is true
(Ee t+ 1 / Et) = (1 + R t, $)/ (1 + R t, GBP)
And
(E et+ I - Et)/Et = (Rt, $ - Rt, GBP)/(1 + R t, GBP ) ≈ (Rt, $ - Rt, GBP)
If UIP is true, the interest rate differential between two countries is
approximately equal to the expected rate of depreciation of the high
interest rate country!
Under UIP, realized nominal returns across countries may not be equal,
but expected returns are. This means that interest differentials reflect the
expected rate of exchange rate depreciation, with high interest rate
countries on average depreciating against low interest rate countries.
89
Implication
If UIP holds, an investor can't make money on average by investing in
foreign currency bonds with interest rates that are higher than
domestic interest rates.
This is because the higher foreign interest rate is, by UIP, on average
offset on average by the depreciation of the foreign currency relative
to the home currency.
Investors who care only about expected returns (not variances) will
bid up the price of foreign currencies with high interest rates until
they are expected to depreciate!
If such investors are decisive (and have enough capital), UIP will hold.
Whether or not UIP holds is an empirical question.
If UIP does not hold, there is an expected profit to buying foreign
currency of countries with high interest rates, but it is risky.
There is no free lunch in international finance!
90
Covered Interest Parity:
Do the same investment in GBP deposit, but
take out all the risk by selling the known GBP proceeds forward
for USD in the forward market.
I agree with counterparty today to exchange known quantity of GBP 1 year
forward for known quantity of dollars. The rate of exchange is the forward rate
Ft,1.
Known USD proceeds in 1 year with GBP investment and forward sale of
proceeds
(1 + R t, GBP)(F t, 1 / Et)
Known USD proceeds to investing in USD deposit:
(1 + R t, $).
Under Covered Interest Parity
(1 + R t, GBP )(Ft, 1 / Et) = (1 + Rt, $)
91
Thus, the forward exchange rate is pinned down by the spot rate, and home
and foreign interest rates!
Moreover, if UIP also holds, since CIP always holds , we have
Ft,1 = Eet,1
That is , under UIP the forward exchange rate is equal to the expected future
spot exchange rate!
Combining UIP with CIP we find
Ft.N= Eet+N
The forward rate is an efficient predictor of the spot rate.
92
Uncovered Interest Parity in Real Terms
In nominal terms UIP says
(Ee t+ 1 - Et)/Et ≈ (R t, $ - Rt, GBP)
Subtract expected inflation differentials from both sides
∆Ee t+1/Et – (πet,1 – π*et,1) = rre t,1 – rr*e t, 1 = ∆Qe t+1/Q t
If ex ante relative PPP is also true, the left hand side equals 0 and the
expected change in the real exchange rate is 0. THIS MEANS THAT UIP
AND RELATIVE PPP TOGETHER IMPLY THAT REAL INTEREST RATES
ARE EQUALIZED EX ANTE PERIOD BY PERIOD!
Under UIP and Relative PPP, realized real returns across countries may
not be equal, but expected real returns are. This is true regardless of
monetary policy or real demand and supply shocks.
A very strong condition. Is it true?
93
Cumby and Obstfeld proposed a simple test under the
maintained hypothesis of rational expectations. If ex ante real
rates are equalized
(πet,1 – π*et,1) = Rt,1 – R*t,1
Under rational expectations,
(πet,1 – π*et,1) + u t,1 – u* t,1 = (πt,1 – π*t,1)
Where the u’s are forecast errors orthogonal to time t
information. Which Implies
(π t,1 – π* t,1) = Rt,1 – R*t,1 + u t,1 – u* t,1
So a regression
(π t,1 – π* t,1) = α+ β (Rt,1 – R*t,1 ) + u t,1 – u* t,1
Should produce α= 0 and β = 1. If β = 1 and a non zero, then
ex ante real rates not equal but differ by a constant.
94
Cumby and Obstfeld proposed a simple test under the
maintained hypothesis of rational expectations. If ex ante real
rates are equalized
(πet,1 – π*et,1) = Rt,1 – R*t,1
Under rational expectations,
(πet,1 – π*et,1) + u t,1 – u* t,1 = (πt,1 – π*t,1)
Where the u’s are forecast errors orthogonal to time t
information. Which Implies
(π t,1 – π* t,1) = Rt,1 – R*t,1 + u t,1 – u* t,1
So a regression
(π t,1 – π* t,1) = α+ β (Rt,1 – R*t,1 ) + u t,1 – u* t,1
Should produce α = 0 and β = 1. If β = 1 and α non zero,
then ex ante real rates not equal but differ by a constant.
95
SPECIAL TOPIC: CURRENCY Basis
I am a Europe bank and I agree today t to SELL 100 Euro forward in 3 months at the forward rate F(t,90).
By doing this I am LONG USD (in 90 days) and SHORT Eur (in 90 days). I want to HEDGE my exposure
by BUYING Euro spot (a long position) and BORROWING USD (a short position).
How much do I buy today t? 100/(1 + R(t,90)_eur_l) Euros
I finance this by borrowing today time t 100E(t)/(1 + R(t,90)_eur_l) dollars.
In 90 day t+90
I deliver 100 Euros to counterpart
I receive 100*F(t,90) dollars
I pay off my USD loan 100E(t){(1+R(t,90)_usd_b/(1 + R(t,90)_eur_l) dollars
Final cash flow at t+90 = 100*F(t,90) - 100E(t){(1+R(t,90)_usd_b/(1 + R(t,90)_eur_l) dollars. Riskless
cash flow is 0 if and only if
F(t,90)/E(t) = (1+R(t,90)_usd_b/(1 + R(t,90)_eur_l)
So if European banks are making the market to provide Euros forward, the forward exchange rate
will reflect their marginal cost od USD funding. Since 2008, the marginal cost of USD funding
reflected in forward rates has been higher than reported Libor rates.
Libor is unsecured bank funding. Alternative is to fund forward rates with secured repo collateral
(Europe banks pledge USD assets against borrowing).
96
• “…all transactions associated with the change of ownership in external financial
assets and liabilities of an economy”
Definition and Types
Savings, Investment, Current Account
Basic National Income Identities
GDP = C + I + G ~ closed
GDP= C + I + G + EX -IM ~ open
GDP is value of goods and services produced in
US.
GNP is value of goods and services produced by
US workers and US owned factories
GNP = GDP + Net Foreign Factor Income
The current account, CA, is the sum of the trade
balance, EX - IM, net factor income from abroad
CA = EX - IM + Net Foreign Factor Income
98
Thus
GNP= C + I + G + CA
Define National Saving as
SN = GNP - C - G
Subtracting C and G from both sides of
the national income identity,
We obtain the key equation of
international finance
SN - I = CA
99
Thus, as matter of accounting, any country that
runs a current account deficit is a county in which
national saving falls short of domestic Investment.
Any country that runs a current account surplus is
a country in which national saving exceeds
domestic investment.
We gain additional insight by noting that
SN = (GNP - T - C) + (T - G)
SN = SP + SG
(SP - I) + SG = CA
100
In an open economy, saving need not equal investment.
Rather we have
SN - I = Net Capital Outflow
Thus, by accounting, a country's net capital outflow must equal
its current account surplus
CA Surplus (Deficit) = Net Capital Outflow (Inflow)
The excess of national saving over investment is the net. outflow
of loanable funds abroad
Over the past 25 years, the US has been a large international
borrower (net capital inflow to US)
Japan has been a large international lender (net capital outflow
from Japan).
101
Saving Investment and the US Current Account 1960- 2015
102
A country running a current account deficit means
There is an excess of domestic investment over
national saving
Country must be financing this deficit with a capital
inflow - borrowing from abroad.
We can further break down these capital flows into
private and official (central bank) flows
CA = Private Capital flow + Official Capital Flow
103
Balance of Payments Accounts
104
Current Account Transactions
In Millions of USD
Receipts
Payments
Exports (line 2)
2,279
Imports (line 10)
2,758
Income
Received
(line 5)
794
Income Paid
(line 13)
570
Transfer
received (line 8)
126
Net Transfers
Paid (line 16)
249
Current
Account Deficit
376
105
Capital Account Transactions
In Millions of USD
Receipts
US Assets
Purchased by
Foreign
investors (line
24)
1,041
US Reserve
Assets (line 23)
3
Net Capital
Inflow (line 24 –
line 19- line 28)
Payments
Foreign Assets
Purchased by
US Private
Sector (line 19line 23)
646
Net Financial
Derivatives (line
28)
2
Statistical
Discrepancy
19
395
106
US Net International Investment Position and Exorbitant Privilege
The US has run CA deficits for 30 years so has a large net international liability
position with ROW.
Because of its ‘exorbitant privilege’ as the provider of the global reserve currency,
the U.S. reaps a capital gain when the dollar depreciates, since U.S. assets abroad
are mostly foreign currency denominated while US liabilities owed to foreign
investors are almost entirely dollar denominated.
Thus, an orderly decline in the dollar can facilitate global portfolio adjustment by
reducing the value of US net international liability position, so long as the US retains
the privilege.
However, there is ultimately “no free lunch” for the U.S. from dollar deprecation.
Eventually, a weaker dollar will worsen the U.S. terms of trade, slowing growth of
U.S. living standards and, ultimately, U.S. demand.
The US has a privilege in the sense that most countries with large net international
liability position are forced by the capital markets to issue liabilities in foreign
currency. As such home currency value of net liability position worsens when their
home currency depreciates.
107
Background
As a matter of accounting, the current account (CA) imbalance must equal the
difference between national saving and investment (I). National saving, in turn, is the sum
of private saving (SP) by households and corporations and saving by the government, or
taxes (T) minus government spending (G).
CA = (T – G) + S private - I
The U.S. runs a current account deficit of roughly 3.5 percent of GDP. We
account for this as follows. First, the government is running a massive budget deficit,
where T-G is negative and subtracts from national saving. The current account deficit is
smaller than the budget deficit because of a surplus of private saving – both household
and business saving (corporate profits) – relative to a depressed level of business and
residential investment.
In textbooks, it is often assumed that the change in the net international
investment position of a country is just equal to the current account balance:
∆NIIP = CA
108
Thus, if the U.S. runs a current account deficit of -$617 billion as it did by
one measure in 2007, the first year of the crisis, the textbook would expect the U.S.
net international investment position to deteriorate by -$660 billion that year.
But as shown in Table 1, it did not. This would only be true if asset prices in
local currency terms are unchanged and if exchange rates are unchanged. In the real
world, asset prices and exchange rates do change and as we see, these have a large
impact on the U.S. international investment position.
Moreover, the impact of asset price and exchange rate changes on the net
international investment position depends on the size, composition, and currency
denomination of the gross holdings of U.S. assets abroad and foreign claims against
the U.S. US assets abroad are tilted toward equity and FDI while foreign claims
against US are tilted toward fixed income. Thus in the real world we have:
∆NIIP = CA + (effect of asset price changes local currency)
+ (effect of currency changes)
109
US Net International Investment Position
Between 2002
and 2009, US
Net International
Liability was
virtually constant
at roughly 2.5
trillion dollars.
Yet during that
time US ran
cumulative
current account
deficits of more
than 4.5 trillion
dollars!
How to reconcile
-different
portfolio mix
-US Liabilities in
dollars so
weaker USD
improves NIIP!
The numbers
are big and a
function of gross
positions
110
111
112
113
114
115
116
International Financial Architecture
Capital Controls
1. Controls on outflows unlikely to work
2. Controls on inflows may prevent lending boom and financial crisis, but cause
distortions
Role of IMF
1. There is a need for international lender of last resort (ILLR) and IMF has
played this role
2. ILLR creates moral hazard problem
3. IMF needs to limit moral hazard
Lend only to countries with good bank supervision
4. Need to do ILLR role fast and infrequently
Sudden Stop
• A sudden stop in capital flows is defined as a sudden slowdown in private capital
inflows into emerging market economies, and a corresponding sharp reversal
from large current account deficits into smaller deficits or small
surpluses.[1] Sudden stops are usually followed by a sharp decrease
in output, private spending and credit to the private sector, and real exchange
rate depreciation. The term “sudden stop” was inspired by a banker’s comment
on a paper by Rüdiger Dornbusch and Alejandro Werner about Mexico, that “it is
not speed that kills, it is the sudden stop”.[2][3]
• Sudden stops are commonly described as periods that contain at least one
observation where the year-on-year fall in capital flows lies at least two standard
deviations below its sample mean.[4] The start of the sudden stop period is
determined by the first time the annual change in capital flows falls one standard
deviation below the mean and the end of the sudden stop period is determined
once the annual change in capital flows exceeds one standard deviation below its
sample mean.
Charts on Monetary Model
119
Geometry of Monetary Model for special case λ = 1
δ
E
zf/E =
EE
Slope = z
We have used the fact
that we can turn an
infinite horizon model
into a 2 period model if
the (composite) forcing
variable is expected to
revert to a random walk
in the next period. Note
that if λ > 1 EE is
convex not linear but
analysis will still go
through. In all future
period i = i* = rr an
exogenous world real
interest rate. It is this
exogenous long run
real interest rate that
pins down the level of
future prices and thus
the level of current
prices.
Assume P
and P* are prices of
traded
goods
and
Q=1..
IP
1
120
Temporary Present Rise in M or Fall in M*
Exchange Rate Depreciates
Interest Rate Differential Declines
E
zf/E
EE
Slope = z
We also know that P/P*
rises in the present since
by relative PPP EP*/P = 1
is constant.
We also
know that with UIP and
relative PPP real interest
rates are equalized so
expected and actual home
inflation between present
and future falls.
IP
121
Temporary Present Rise in Y or Fall in Y*
Exchange Rate Appreciates
Interest Rate Differential Rises
E
zf/E
EE
Slope = z
We also know that P/P*
falls in the present since
by relative PPP EP*/P = 1
is constant and assumed
exogenous.
Thus
expected and realized
home
inflation
rises
relative to foreign inflation.
IP
122
Future Permanent Rise in M or Fall in M*
Exchange Rate Depreciates
Interest Rate Differential Rises with Expected Inflation
zf (M’) /E
E
zf (M) /E
EE
We also know that P/P*
rises in the present and
future since by relative
PPP EP*/P = 1 is
constant.
Under UIP
and Relative PPP real
interest
raters
are
equalized so there must
be
inflation
between
present and future.
Slope = z
IP
123
Future Permanent Rise in Y or Fall in Y*
Exchange Rate Appreciates
Interest Rate Differential Falls with Expected Disinflation
E
zf (Y’) /E
zf/E
EE
We also know that P/P*
fall in the present and
future since by relative
PPP EP*/P = 1 is
constant.
Under UIP
and Relative PPP real
interest
raters
are
equalized so there must
be disinflation between
present and future.
Slope = z
IP
124
Future Permanent Rise in M or Fall in M* - Present
Exchange Rate Depreciates in the Present
Interest Rate Differential Rises in the Present
E
EE
We also know that P/P*
rises in the present and
since by relative PPP
EP*/P = 1 is unchanged in
the present..
Slope = z
IP
125
Future Permanent Rise in M or Fall in M* - Future
Exchange Rate Depreciates in the Future
Interest Rate Differential Returns to 1
E
EE
Slope = z
IP
1
126
MacDonald and Taylor: Asset Market Approach Implies Co Integration, Cross
Equation Restrictions, and Error Correction
In their notation st is spot rate and xt = (mt – m*t) – γ(y t – y*t)
Solve this forward
Also note that
127
Subtract xt from both sides
st – xt =
– xt
Which can always be simplified to
Now st as a forward looking asset price exhibits unit root/near random
walk behavior. From this equation we see that spot rate is co integrated
with composite fundamental xt which also has unit root.
The theory (Campbell and Shiller) says that the equilibrium error Lt is
the best available forecast the dpv of the growth in fundamentals. This
implies cross equation restrictions on a VAR model of ∆xt and Lt.
128
Define z t
And write VAR as
Then Lt must equal
With h’ = [1 0]’ and g’ = [0 1]’. This infinite dpv has a convenient closed
form
Lt = g’ z t
With
129
Post multiplying we see that the restrictions can be written as
Term by Term this implies [-ψa21 1 – ψa22] = [ψa11 ψa12]
So even in a simple VAR(1) model there are testable restrictions. Note to
implement need to take a stand on income elasticity of money demand γ as
well as the discount rate ψ. Money demand elasticity can be estimated
from co integrating regression or imposed as 1. Most researchers impose
discount rate. However note this is not necessary as it can be estimated
under the restrictions.
Bottom line is that these restrictions are rejected (as they usually are in these
models). Moreover the actual Lt equilibrium error is much more volatile than
theory implies given the forecast ability of future changes in xt.
130
131
Rational bubbles in Asset Market Model
132
Some Micro Foundations for Monetary Approach via Obstfeld Rogoff
Subject to
With PT,t = Et and with Bt denoting bond holdings indexed to traded goods
inflation
Optimal Money Demand in O.R. model
Let U = CTγ CN1-γ denote exact consumption index and note that 1 + iT,t+1 =
(1 + r)PT,t+1/PT,t is the realized nominal return on an indexed bond. We then
have
1
1



Mt  U 
U  





i
Pt
 i 
 1  i 
So demand for real money balances is log linear in the exact consumption
index and linear in the nominal return on bonds and is deflated by consumer
price index.
Micro Foundations for the Monetary Model
Consider an extension of OR with a world of a large number of economies, two
of which are ‘home’ and ‘foreign’. The real interest rate is pinned down in rest
of world at r and the ROW nominal price of the traded good is constant and
equal to 1.
Let E denote home currency price of ROW currency and E* foreign price of
ROW currency. Allow consumers in home to invest in inflation indexed bond
indexed to E = PT consumers in foreign to invest in inflation indexed bonds
indexed to E* = P*T.
Note that if foreseen ‘shocks’ to traded endowment are permanent then in
equilibrium there will be no trading in bonds as trade will be balanced period
by period so long inflation indexed bond prices consumers confront are as
above.
1
1



M t  U 

and

Pt  i
 1  i 
M *t  U * 


i
*
P *t 
 1  i * 

These can be re written as
1
 U 



 and
i
 1  i 

Mt
E 1 t
M *t
E * *1 t
 U * 




i
*
 1  i * 
1

Where η = PN /PT and by triangular arbitrage EH,F = E/E*. So pairwise the
nominal exchange rate between any two countries is determined by
1/ 
E
H ,F
M U *



M * U 
1
 * 
 
 
1/ 
1/ 
 i  1 i * 
  

 i *   1 i 
This is a close cousin of the old monetary model of exchange rates.
Lets approximate for the case that (1+i)/(1+i*)≈1.
Lets approximate for the case that (1+i)/(1+i*)≈1.
1/ 
E
H ,F
M U *



M * U 
1
 * 
 
 
1/ 
 i i*
1 

i* 

Taking logs
e
H ,F
1
1
 m  m *  (u * u )  (1   )( *  )  (i  i*)

r
Where we evaluate the log linearization at i* = r. In the special case where
unforeseen shocks to traded output are assumed to be permanent ex post
u = y and u* = y* where y and y* are real GDP deflated by the exact
consumption price index
e
H ,F
1
1
 m  m *  ( y *  y )  (1   )( *  )  (i  i*)

r
Evaluating Approximation for quarterly i = 0.011, i* = .01,r = .01, and ε = 2
1/ 
 i 
ln  
 i*
1/ 
1 i * 
 ln 

 1 i 
0.047 ≈
1
 (i  i*)
r
.05
So since this works off the first order conditions it will hold as a structural
equation regardless of (non traded) goods prices being sticky or flexible.
Note that it does impose law of one price for traded goods.
Monetary Approach to Exchange Rates with Taylor Rule Central Banks
Monetary approach developed in the 1970s at Chicago when paradigm was to
think of central banks as setting a path for mt. Last 15 years, it is recognized that
central banks set feedback (Taylor rules) for nominal interest rates and that money
supply is endogenous, not a control variable. Fortunately the logic of the monetary
approach continues to apply and in a very elegant way with TR central banks. Start
with UIP in real terms (we will later discuss how to add a risk premium)
rre t,1 – rr*e t, 1 = ∆Qe t+1/Q t
And approximate the RHS as qe t,t+1 – q t. We have
q t = qe t,t+1 + rr*e t,1 – rre t, 1
Assume that relative PPP holds in the long run so that q t is a strictly
stationary process with unconditional mean q. Solving forward we have
q t = q + Et ∑i=0,∞ (rr* t+i,1 – rr t+i, 1)
Suppose home central bank sets policy rate according to
Rcb t = rr + πT + 1.5(π e t,1 – πT) + 0.5(y t – y p t)
139
Nominal Exchange Rate as an
Instrument under IT
140
E. Using NER as a policy instrument
 Recall the monetary conditions index
mcit  b4 rˆt  (1  b4 )(  zˆt )
RIRate gap
RER gap
zt  st  pt*  pt
rt  it  Et { t 1}
zˆt  zt  zt
rˆt  rt  rt
 So far, we assumed that independent monetary policy (e.g., IT) is implemented
via the nominal interest rate…
 … and via affecting the real interest rate immediately
 We can do that by manipulating the nominal exchange rate!
 and immediately affecting real exchange rate instead
 So, the NER becomes a POLICY INSTRUMENT!
141
E. Using NER as a policy instrument
 … model implementation
 the CB intervenes to achieve a “desired” rate of ER change
 …. BUT, not just to smooth the ER fluctuations, rather to fulfill the target
rate of ER change as is set by the policy rule for ER.
 The policy rule for the targeted change in the ER:
stT  f1stT1  (1  f1 )(stT _ Neutral  f 2 ( Et t 3   T )  f 3 yˆ t )   ti
 “Neutral” change is consistent with inflation differential and the change in
equilibrium RER:
s T _ Neutral   T   *  z
t
t
t
t
142
Main objectives of Fiscal Policy
•Stabilization
•Allocation
•Distribution
143
Stabilization
• Using fiscal policy to smooth fluctuations in output
• Budget balance increases when output rises and decreases when it
falls
• Issues:
• Fiscal space to respond to changes in output
• Fiscal framework and degree of automatic stabilizers
• Fiscal multipliers
• Sustainability
144
Allocation
• Ensuring spending is allocated toward
long term development priorities
• Transport, education, etc.
• Both across sectors and within
sectors (roads, ports, etc.)
• With sufficient capacity to respond to short
term objectives like stabilization
• Issues:
• Fiscal adequacy (Fiscal space)
• Fiscal framework
• Fiscal effectiveness and efficiency
145
Distribution
• Using spending, taxes and transfers to have
an impact on the distribution of income
throughout the country
• Transfer mechanisms: public pensions,
unemployment assistance, welfare, etc.
• Spending mechanisms: pre-school education,
wage subsidies, employment training, etc.
• Tax mechanisms: Progressive income taxes, Low
income tax credit, etc.
146
Kinds of questions economist will try to answer in regular work
Macro focused fiscal questions
• Is fiscal policy sustainable?
• What is the size of the fiscal stimulus needed to achieve short term growth acceleration of
2%?
• How has the fiscal stance changed to accommodate falling revenues?
Public finance focused questions
• How can a country improve the ability for its public finance system to mitigate inequality?
• How should government expand its domestic revenue?
• How can country improve the efficiency of its spending in the health sector?
• How should country prioritize fiscal consolidation?
• How can countries’s fiscal framework be improved to mitigate the impacts of external
shocks?
• How can the country’s intergovernmental fiscal framework be improved?
147
Core macro focused fiscal analysis
• Analysis of developments in fiscal outcomes (deficit, etc.), and
sources underlying these developments
• Assessment of spending and broad outcomes (comparators)
• Assessment of revenues/GDP (comparators)
• Assessment of fiscal space
• Assessment of fiscal sustainability
• Analysis of sources of fiscal risks
• (After the crisis): Estimations of fiscal multipliers
148
Fiscal stance
• What: Assessment of fiscal policy stance, fiscal sustainability, and
discretionary changes in fiscal policy
• Expansionary or restrictive?
• Pro-cyclical?
• How: Cyclically adjusted or structurally adjusted
• Issues: Fiscal balance not just the result of actual government decisions.
Also dependent on business cycle, windfall revenues, changing
asset/commodity prices
149
Fiscal sustainability
• What: Analysis to ensure that fiscal policy framework can be sustained:
• Will not result in explosive debt
• Will not create financing needs that can’t be met by resources available to the
public sector
but also…
• Has sufficient ability to adjust public spending to absorb shocks (stabilization)
• Is inclusive of contingent liabilities
• Solvency and liquidity
• Debt stabilizing primary balance:
150
What can undermine fiscal sustainability?
• Things that impact the ability to service debt:
• Debt structure and composition
• Shocks (interest rates, exchange rates, economic growth,
exports, domestic revenue)
• Unforeseen borrowing/contingencies
• Focus on:
• Stress testing debt profile (DSA)
• Minimizing exposure to shocks
• Minimizing unforeseen outlays
• Rules-based fiscal frameworks
• Accounting for fiscal risks
151
Sustainability may call for rules based fiscal frameworks
• Discretionary fiscal policy optimal for macroeconomic stabilization
but
• Policy failures abound in practice
• Time inconsistency, Common pool problems, Deficit bias, Procyclical bias, Expenditure
composition bias, Optimal forecast bias
• Consequences
• Macro instability, Fiscal sustainability problems, Reputational cots, Vulnerability to
shocks/Sudden stops
• What rules based fiscal frameworks aim to do:
•
•
•
•
Fiscal authorities commit to pursue a predictable, transparent, fiscal policy course
Well defined constraints
Constrained discretion
Guided by good practice
152
Fiscal Multipliers
• When they’re important:
• Economic downturn, countries implement fiscal stimulus to
cushion the impact
• Fiscal deficit: need to undertake fiscal consolidation
• What it might look like:
•
•
•
•
•
Credit lines to private sector
Lump sum payments to retirees
Reduction in taxes
Program of public works
Tax moratorium
• Why you need to estimate: to know what kind the stimulus the government would
need to implement for a given short term growth response (or likely impact of a
fiscal cut)
• Problem: Hard to estimate because of endogeneity
153
Fiscal Multipliers
• Estimation techniques:
• Macroeconomic forecasting models: Assuming
historical relationships
• Time series models: Usually, structural VARs with
timing assumption…need high frequency data
• DSGE models
• For countries with less data: “Bucket approach” –
grouping countries into groups likely to have similar
multiplier values based on their characteristics
154
What we know about Fiscal
Multipliers
Trade
Openness
Open
Exchange rate
Fixed
< Closed
> Flexible
Business Cycle
Full
Recession >
Debt
Automatic
stabilizers
<
Strong <
Expend/tax
mgmt
Strong
High
employment
Low
Weak
> Weak
155
Core public finance oriented analysis
• Allocations of public spending versus development
objectives
• Spending versus outcomes (effectiveness,
efficiency)
• Distributional analysis of spending, taxing (and
changes in taxes/spending)
• Analysis of sources of fiscal space:
• Analysis of expanding revenue sources
• Analysis of spending inefficiencies
156
Expanding fiscal space
Tax Revenue
1.0
0.8
0.6
Fiscal
Diamond
0.4
0.2
Expenditure
efficiency 0.0
Borrowing
0.2
0.4
0.6
0.4
0.2
Aid
0.0
0.2
0.4
157
Fiscal space: Improving
Fiscal Efficiency
• What: (Generally) Analysis of current public spending
and services provided with a view to decreasing the
inputs while maintaining same service provision, or,
alternatively, increasing outputs with the same level of
inputs (technical efficiency).
• What can derail it:
• Poor execution (leakage, lags)
• Lack of coordination (duplication)
• Weak institutions/poor policies/outdated systems and
processes)
• Fiscal framework (for example, fiscal devolution)
• Etc.
158
Download