Dollarization and Crises: Ways In and Out

1
Dollarization and CrisesWays In and Out

• Alejandro Izquierdo
• De-dollarization Strategies and Domestic Currency
  Debt Markets in Emerging Economies
• Okinawa, Japan
• April 8 2005

2
Outline

• Sudden Stop, Devaluation and Dollarization Key
  Facts
• Determinants of Sudden Stops Domestic Liability
  Dollarization
• How did we get there?
• Ways out Successful Experiences

3
Sudden Stop in Emerging Asia and LAC-7 (Capital
Flows, USD million, last four quarters)
Asian Crisis
Russian Crisis
LAC-7
Millions of USD
Emerging Asia
Note LAC-7 includes Argentina, Brazil, Chile,
Colombia, Mexico, Peru and Venezuela. Emerging
Asia includes Indonesia, Korea, Malaysia,
Philippines and Thailand. Source Central Banks.
4
Real Exchange Rate Adjustment
(vis-à-vis US dollar, Jan-90100)
210
Russian Crisis
190
Emerging Asia
170
150
130
110
LAC-7
90
70
Jan-90
Jan-91
Jan-92
Jan-93
Jan-94
Jan-95
Jan-96
Jan-97
Jan-98
Jan-99
Jan-00
Jan-01
Jan-02
Jan-03
Jan-04
Note LAC-7 includes Argentina, Brazil, Chile,
Colombia, Mexico, Peru and Venezuela. Emerging
Asia includes Indonesia, Korea, Malaysia,
Philippines and Thailand.
5
Domestic Liability Dollarization

• For some countries, credit in dollars was high as
  share of GDP (Domestic Liability Dollarization,
  or DLD) before the crisis
• Payments system is at stake following devaluation

Source Calvo, Izquierdo and Mejia (2004) and own
calculations.
6
Outline

• Sudden Stops and Devaluation Key Facts
• Determinants of Sudden Stops Domestic Liability
  Dollarization
• How did we get there?
• Ways out Successful Experiences

7
Determinants of Sudden Stops DLD and Tradable
Output

• Low tradable output levels (as a share of the
  absorption of tradables) will place the brunt of
  adjustment on the real exchange rate (RER) if a
  Sudden Stop occurs.
• Together with high DLD they are an explosive
  cocktail

8
Tradable Output

• Key difference between Emerging Asia and LAC-7
  Tradable output (Y) relative to the absorption
  of tradable goods (A) (compensates for higher
  pre-crisis dollarization)
• Emerging Asias reaction to currency depreciation
  was stronger in terms of the expansion of
  tradable output

1.30
1.25
Emerging Asia
1.20
1.15
1.10
1.05
1.00
0.95
LAC-7
0.90
0.85
0.80
1996
1997
1998
1999
2000
2001
2002
2003
EA Pre-crisis ? 0.93. LAC-7 Pre-crisis ? 0.88
Source Own calculations based on data from WDI
and WEO databases.
9
Dollarization is an Addiction,Just Like Smoking

• We know that dollarization is bad for a countrys
  health it brings crisis (cancer)
• As with any addiction, the first step is to
  acknowledge that there is a problem. But in
  order to quit, two questions must be answered

• How do countries get hooked on smoking?
• Is there a patch to stop smoking?

10
Outline

• Sudden Stops, Devaluation and Dollarization Key
  Facts
• Determinants of Sudden Stops Domestic Liability
  Dollarization
• How did we get there?
• Ways out Successful Experiences

11
Latin America Macroeconomic Policies and
Currency Substitution

• A history of high fiscal deficits, loose monetary
  policy and high inflation lies behind most
  dollarization episodes in LAC

12
Domestic Liability Dollarization

• Spreading the virus to the financial sector
• Banks have typically matched dollar deposits with
  dollar loans
• But a large share of dollar loans was on lent at
  home, leading to currency mismatches in
  non-tradable sectors and balance-sheet effects

13
Emerging Asia Bank Foreign Borrowing (Foreign
Liabilities / (Total Deposits Foreign
Liabilities))
Note Average for Indonesia, Korea, Malaysia,
Philippines and Thailand. Source IFS
14
Outline

• Sudden Stops, Devaluation and Dollarization Key
  Facts
• Determinants of Sudden Stops Domestic Liability
  Dollarization
• How did we get there?
• Ways out

15
Latin America Currency Substitutionand
Hysteresis

• Reducing inflation (the factor that made
  countries engage in DLD) should be part of the
  strategy. But is it enough?
• Hysteresis cumulative experience in using
  dollars reduces the cost of transactions in
  dollars (Uribe, 1997). Reducing inflation is not
  enough to reduce dollarization

16
What Didnt Work

• Reinhart, Rogoff and Savastano (2003)
• Forced de-dollarization does not seem to be the
  answer) Bolivia and Peru
• Even when forced de-dollarization succeded, there
  were costs
• The size of the domestic banking system fell
  (credit to the private sector in Mexico halved
  two years after conversion in 1982)
• Capital flight substantial holdings of offshore
  deposits
• Taxes on holdings of US deposits leads to
  disintermediation via off-shore deposits (Peru)

17
What Did Work The case of Israel

• Reinhart et al (2003) identify only two
  successful cases of lasting declines in
  dollarization without heavy costs in financial
  intermediation or capital flight Israel and
  Poland
• Flexible exchange rates have been proposed as the
  main factor leading to lower liability
  dollarization.
• But the choice of a fixed exchange rate regime
  may be the consequence rather than the cause for
  liability dollarization, because the costs of
  floating are high when heavily dollarized.
• How did Israel do it? Were there any additional
  patches?

18
Inflation RER Volatility

• If individuals care about buying a basket of
  goods, they will allocate their savings so that
  they minimize the risk of being unable to buy
  that basket.
• The return on dollar deposits depends on the
  dollar price of that basket (RER depreciation)
• While the return on domestic-currency deposits
  depends on the peso price of that basket
  (inflation)
• Low volatility of inflation relative to the
  volatility of RER depreciation should lead to
  de-dollarization.

19
Israel Inflation and RER Volatility
Inflation Volatility
Real Depreciation Volatility
Note Variances are calculated over a 5-year
moving window. Source Own calculations based on
data from IMF-IFS.

• Lowering inflation volatility relative to real
  exchange rate volatility may be part of the answer

20
Israels De-dollarization and Exchange Rate Band
21
Or was it Also the Patchesor Additional
Measures?

• But Israel also pursued additional policies
  (Galindo and Leiderman (2003)), many of them
  patches to reduce the costs of floating
• Initially, one year mandatory holding period for
  dollar deposits
• Offered CPI-indexed deposits
• Banks required active hedging of currency risk
  for non-tradable activities
• Active development of financial derivatives
  markets
• Made effort to deepen local currency bond markets
• And it worked deposit dollarization went down to
  18 of total deposits (2001) from 45 (1985)

22
The Current Situation Long-run Trend or
Short-run Opportunism?

• Renewed interest in domestic currency lending
  (e.g., the case of Colombia, 2004)
• Is it Leaning against the wind policies and
  appreciation expectations? Or,
• Based on the Argentine experience can it be more
  costly to lend in dollars?

23
The Role of IFIs?

• Multilateral lending in domestic currency,
  hedging currency risk with the recipient country
  To what extent does this shield EMs from the
  effects of dollarization?
• Issuance of domestic-currency multilateral debt
  in the recipient country (on lent in domestic
  currency) Will it crowd out issuance of public
  debt? (country risk vs. exchange rate risk).

24
Dollarization and CrisesWays In and Out

• Alejandro Izquierdo
• De-dollarization Strategies and Domestic Currency
  Debt Markets in Emerging Economies
• Okinawa, Japan
• April 8 2005