TAUFIK (43714827)

Word Count: 875 (excluding reference, tables, and graphs)

Question E1
Based on Eun and Resnick (2015), the arbitrage opportunity in (b) known as the Covered
Interest Arbitrage (CIA) can be captured when interest rate parity (IRP) is violated, in which the
actual forward rate is different with implied forward rate:
F S x (1+i_L) (1+i_H)
The data in (b) revealed that only 4 (four) period in which the covered interest arbitrage can
be captured, using the following strategy:
F/S (1+i_L)/(1+i_H) Ft (L$/H$) FL$/H$
Time (1) - (2) (3) - (4) Conclusion
(1) (2) (3) (4)
12/30/1974 0.9930 0.9931 (0.000090) 23.2339 23.2360 (0.0021) IRP doesn't hold; Sell L$, Buy H$ in Forward Market
12/30/1983 0.9713 0.9713 (0.000037) 21.2251 21.2259 (0.0008) IRP doesn't hold; Sell L$, Buy H$ in Forward Market
12/30/1993 0.9683 0.9681 0.000208 20.1467 20.1424 0.0043 IRP doesn't hold; Sell H$, Buy L$ in Forward Market
12/30/2001 0.9739 0.9738 0.000046 18.0187 18.0178 0.0009 IRP doesn't hold; Sell H$, Buy L$ in Forward Market

Based on the common profit generated strategy of buy low - sell high, when IRP did not
hold, the appropriate strategy is selling the overvalue currency to get/buy the undervalue currency.
However, we can also see that the differences between the actual Forward rate (Ft) and the implied
Forward rate (FImplied) at those periods were not significant, the CIA may not work well since the
generated profit might be ruled out by all relevant transaction costs.

Question F
The Forward premium under the expectation hypothesis is the expected rate of depreciation
(Backus, Foresi, and Telmer, 1995). Therefore, the standard regression test used in excel file (c) is
based on the ex-post depreciation in future spot rate on the current forward premium (Wu and
Zhang, 1997), that is:

From the data in (c), the forward premium/discount can only explain 7.6% variation (R2) of
the changes in future spot rates. The negative intercept (-0.44) suggesting that the forward rate
premium/discount tends to overestimate the changes in the future exchange rate (Ott and
Veugelers, 1986). This finding also supports the subsample study of Chiang (1988) that the joint-
rolling regression reject the unbiasedness hypothesis in many cases.

1
TAUFIK (43714827)

The data in (c) reveal the residual output of the regression that can be traced to explain the
bias. These outliers allow us to recognize the idiosyncratic events in certain periods which may
cause the unpredicted movement in the future exchange rates that lead to significant bias.

Question G
The possibility to gain from UIA tends to be higher than CIA since the uncovered interest
rate parity (UIP), which stated that interest rate differential between two economies should equal
to the expected rate of change in the exchange rate, often doesnt hold (Eun and Resnick, 2015).
From the data in (b), interest rate disparity exists every year within the 50 years period, hence it
is possible to carry currency trade strategy by borrowing from low yielding currency to fund the
investment in high yielding currency.
However, the carry trade strategy in UIA may not be worked well since there is no absolute
certainty toward the changes in future exchange rate. From the data in (b), out of 49 periods in

2
 TAUFIK (43714827)

which carry trade strategy is conducted to, 22 periods (44.9%) resulted in the loss because the
lower yield currency depreciated more than its interest rate differentials.
Based on Eun and Resnick (2015), the UIA can be captured when i_L - i_B = E(e)L$/H$ 0
as long as E(e)L$/H$ > (St+1-St)/St with L$ as home currency. The following carry trade strategy is:
- Borrow money from the lower-yield currency
- Convert the money to the higher-yield currency
- Invest the money for 1 year at higher interest rate
- Convert the money 1 year later, back to the lower-yield currency
- Pay the debt and earn the arbitrage profit
The carry trade strategy implemented in regards to data (d) are as follows:
1. When interest rate at LotoBoto (i_L) is lower than interest rate at HitoRoof (i_H)
One of the examples is the period of December 1966. The carry trade strategy in this period
are:
(1) Borrow (2) Convert at St (3) Invest 1 year (4) Convert at S1+t Debt to Pay Profit
Time UIA Strategy
H$ L$ L$ H$ H$ H$
12/31/1966 Borrow L$, Invest H$ 1,000,000 20,223,964 21,094,141 969,632 1,052,623 89,272

2. When interest rate at HitoRoof (i_H) is lower than interest rate at LotoBoto (i_L)
One of the examples is the period of December 1968. The carry trade strategy in this period is
the reverse of the first carry trade strategy:
(1) Borrow (2) Convert at St (3) Invest 1 year (4) Convert at S1+t Debt to Pay Profit
Time UIA Strategy
L$ H$ H$ L$ L$ L$
12/31/1968 Borrow H$ Invest L$ 1,000,000 46,418 47,393 988,294 1,022,190 35,018

Variability of Return
The rate of return supposed to be equal to the interest rate disparity if the exchange rate
remains the same (St = St+1), which is rarely exist in the real world. The variability of return in (d)
can be described as follows:
- The Forward was traded at the discount rather than at the premium (36 times and 14 times
respectively), indicating H$ was the higher yield currency.
- The Forward was traded at the discount when L$ has the lower yield than H$, implying that the
market expecting the movement of the future exchange rate to be similar with the interest rate
disparity.

3
 TAUFIK (43714827)

- H$ was tending to depreciate more often than that of L$ (28 and 21 times respectively),
implying that the future exchange rate tends to move in opposite direction and that the currency
risk was discouraging the profit of the carry trade strategy. Therefore, in order for the profit to
be optimized, the final return should be converted to the less depreciated currency, which was
L$.
- Despite the variability of return due to the unpredicted movement in future exchange rate, we
may accumulate profit if we follow the carry trade strategy of borrowing from the low yield
economy and investing in the high yield economy (disregarding the possible movement of
future exchange rate), proven by the accumulated net profit of L$9,954,890 in the case of
LitoBoto and HitoRoof.

Reference

Backus, D., Foresi, S., & Telmer, C. (1995). Interpreting the forward premium anomaly. Canadian
Journal of Economics, Xxviii, 108-119.

Chiang, T. (1988). The Forward Rate as a Predictor of the Future Spot Rate- A. Journal of Money,
Credit, and Banking, 20(2), 212.

Eun, C. (2015). International financial management / Cheol S. Eun, Bruce G. Resnick. (Seventh
ed.). McGraw-Hill/Irwin Education.

Mark, N., & Wu, Y. (1997). Understanding Spot and Forward Exchange Rate Regressions. Journal
of Applied Econometrics (1986-1998), 12(6), 715.

Ott, M. (1986). Forward Exchange Rates in Efficient Markets: The Effects of News and Changes
in Monetary Policy Regimes. Review - Federal Reserve Bank of St. Louis, 68(6), 5.

Yangru, & Hua Zhang. (1997). Forward premiums as unbiased predictors of future currency
depreciation: A non-parametric analysis. Journal of International Money and Finance,
16(4), 609-623.