Lecture 5

Forwards and Futures on interest rates

HKUST, FINA3203, G. Panayotov

 Forward interest rate
So far we assumed that the interest rate is constant. But in fact it is not.

Consider three dates t0 < t1 < t2 , and assume that t0 is today

Let rt0 ,t1 be the continuously compounded (annualised) interest rate from t0 to t1 ,
and rt0 ,t2 be the rate from t0 to t2 .

If you invest $1 today for maturities t1 and t2 , your gross return will be e rt0 ,t1 ×(t1 −t0 )
and e rt0 ,t2 ×(t2 −t0 ) , respectively

We can now infer what the market thinks today of the interest rate from t1 to t2
▶ this ”implicit” rate is called the forward rate from t1 to t2 , denoted ft0 ,t1 ,t2 .
▶ this forward rate is known at time t0 and is expected to be earned from time t1 to t2 .
▶ the forward rate is continuously compounded (and quoted in annualized terms).
▶ e.g. f0,3mth,9mth is the annualized continuously compounded forward rate from three
months from now to nine months from now (recall that today is time 0).

1
By definition: ft0 ,t1 ,t2 = t2 −t1 [rt0 ,t2 × (t2 − t0 ) − rt0 ,t1 × (t1 − t0 )]

HKUST, FINA3203, G. Panayotov

 How to invest at the forward rate?
Suppose you will receive $1 million in cash three months from now. And you are sure
that when you receive it, you will want to invest it for the following six months

Today the three-month T-Bill rate is 3% and the nine-month T-Bill rate is 5%
(continuously compounded)

The six-month T-Bill rate that will be observed in the market three months from now
is not known today. But you know the forward rate.

If you like this forward rate, can you lock it in and thus hedge your interest rate risk?

Problem: you cannot directly invest at the forward rate.

Solutions:
1. synthetic investment at the forward rate
2. Forward rate agreement (FRA)
3. Eurodollar futures (we will not talk about that in this course)
HKUST, FINA3203, G. Panayotov
 Synthetic
3
Borrow today 1 milion × e −0.03× 12 at 3% for three months

3
At the same time, invest this 1 million × e −0.03× 12 at 5% for nine months

Your cash flows:

Cash flow at 0 3mth 9mth

3
borrow at 3% for 3 mths $1 million × e −0.03× 12 −$1 million 0
3 3 9
invest at 5% for 9 mths −$1 million × e −0.03× 12 0 $1 million × e −0.03× 12 e 0.05× 12

total 0 −$1 million 1, 030, 454.53

6
So, you have invested the $1 mln at 6%: $1, 030, 454.53 = $1 million × e 0.06× 12

But also, applying the formula from the previous slide, the forward rate is 6%:
1
ft0 ,t1 ,t2 = 0.75−0.25 [0.05 × 0.75 − 0.03 × 0.25] = 0.06. Mission accomplished.

HKUST, FINA3203, G. Panayotov

 Forward Rate Agreement (FRA)
Most often you would use FRAs instead of creating synthetic investments yourself

FRAs are OTC contracts that can guarantee a borrowing or lending rate
▶ on a given notional principal amount
▶ at a future date
▶ and for a specified period

(rfut −r FRA )×future time period

FRA payoff for the long position: 1+rfut ×future time period × Notional Principle

rfut is the future interest rate that is unknown today. Both rfut and the ”forward price”
(rFRA ) are quoted as an annualized simple interest rate
▶ an FRA at t0 for the future period between t1 and t2 pays off at t1
(rt1 ,t2 −rtFRA
,t ,t )×(t2 −t1 )
▶ this payoff is: 0 1 2
1+rt1 ,t2 ×(t2 −t1 )
× Notional Principle

no arbitrage implies that rtFRA

0 ,t1 ,t2
is equal to the forward rate ft0 ,t1 ,t2 from slide 2 (but
after adjusting for simple vs. continuous compounding!)
HKUST, FINA3203, G. Panayotov
 The example again

You will receive $1 million in cash in three months. You would like to invest it for
another six months after that. You want to hedge the interest rate risk between three
months and nine months from now.

You short today a 3mth − 9mth FRA with notional value $1 mln. Three months from
now the FRA pays off (a positive or negative amount, depending on the future
6-month interest rate ).

You invest that payoff, together with the $1 mln that you will receive in three
months, for the following six months

The interest rate in the FRA is 6.0909% (equal to the forward rate of 6% on slide 4),
but after converting it into a simple rate:
6 6
e 0.06∗ 12 − 1 = X ∗ 12 ⇒ X = 0.060909

HKUST, FINA3203, G. Panayotov

 Cash flows in this example
If the six-month simple interest rate three months from now is 8%:
t=0 3mth 9mth
6
invest $1 mil. at 8% 0 −$1 mil. $1 mil. × (1 + 8% 12 )
= $1, 040, 000
6
(6.0909%−8%) 12 ×$1 mil.
short FRA 0 6
1+8% 12
0
= −$9.178.27
6
borrow at 8% to 0 +$9, 178.27 −$9, 178.27 · (1 + 8% 12 )
pay the loss from FRA = −$9, 545.4
total 0 −$1 mil. $1, 030, 454.6

If the six-month simple interest rate three months from now is 4%

t=0 3mth 9mth
6
invest $1 mil. at 4% 0 −$1 mil. $1 mil. · (1 + 4% 12 )
= $1, 020, 000
6
(6.0909%−4%) 12 ×$1 mil.
short FRA 0 1+4% 126 0
= $10, 249.51
6
invest gain 0 −$10, 249.51 $10, 249.51 · (1 + 4% 12 )
from FRA at 4% = $10, 454.50
total 0 −$1 mil. $1, 030, 454.6
HKUST, FINA3203, G. Panayotov
 Repo transactions
In a repurchase agreement (repo) one party (seller) sells a security with an agreement
to buy it back in the future at a higher price from the other party (buyer). The
security is usually a Treasury bill or bond. The price difference is interest rate.
We can view this as the seller borrowing cash against the security as a collateral.

The repo can also be considered from the perspective of the buyer (or lender of the
cash), and then it is called reverse repo. The buyer buys the security with an
agreement to sell it back in the future at a higher price.
We can view this as the buyer borrowing the security against cash collateral. Then he
can sell the security on the market. At maturity he buys it back from the market,
returns it to the seller, gets back the cash with interest.
HKUST, FINA3203, G. Panayotov
 Why investors use repo or reverse repo transactions?

The seller obtains cash funding

The lender invests essentially risk-free

Financing a long position: purchase a security and enter a repo using the same
security as collateral (this is the same as a long forward WHY?)

Establishing a short position (reverse repo): borrow the security, then sell it, use
the proceeds as collateral in a reverse repo (this is the same as a short forward
WHY?)

HKUST, FINA3203, G. Panayotov

 Collateral and haircuts

The borrower typically also has to post collateral in excess of the notional amount of
the loan - a “haircut”

Haircuts are typically small, but in time of crisis may grow a lot!

HKUST, FINA3203, G. Panayotov

 Example: funding a convergence trade
Newly issued on-the-run 30-year Treasury bonds are typically sold at a slightly lower
yield (and hence higher price) than almost identical off-the-run 29 12 -year Treasury
bonds

Say that we think the spread between these yields has widened too much. Then, with
this trade we bet that the yields of the 30-year and 29 12 -year bonds will converge

This is equivalent to betting that the prices of two bonds will converge.
▶ i.e. the price of 30-year bond will go down relative to the price of the 29 12 -year bond.
The word ”relative” is important - we don’t make any assumption on whether the two
prices will stay near the current levels, or go up or go down!

Arbitrage trade:
▶ Buy off-the-run bond and repo it (to finance our long position)

▶ Borrow on-the-run bond by entering into a reverse repo and sell it (so we establish a
short position)

▶ So, we bought the cheap bond, sold the expensive one, made some money. Very good!

HKUST, FINA3203, G. Panayotov

 10-year off-the-run vs. on-the-run spread

What is a basis point?

HKUST, FINA3203, G. Panayotov

 Case study: LTCM
LTCM was a hedge fund, opened in early 1994. Minimum investment was $10 mln.

At first they did mostly global bond arbitrages, then various equity arbitrage trades.

Relied on understanding of the spreads between yields and prices of various assets –
and hence of sophisticated statistical modeling

Among the founders were some star bond traders and also two Nobel Prize laureates

Strategy: identify small imperfections in the market; exploit them with huge leverage,
secure long-term funding to survive losses, charge hefty fees.

Multiple low-risk arbitrage deals, each with small returns, but on a large scale. Few
directional bets, mostly spread trades with a market-neutral portfolio that gained
value both in rising and falling markets.

You may enjoy reading the book ”When genius failed” by R. Lowenstein, among several
written on LTCM. You can also find a chapter on LTCM in J. Marthinsen’s book that we
mentioned in Lecture 3.
HKUST, FINA3203, G. Panayotov
 LTCM (cont’d)

LTCM did mostly relative value arbitrages, which are based on two principles:

Comovement: Correlations within a pair of assets, A and B, being high.

▶ So that the prices of A and B move together and the difference between their returns
typically does not move much.

Diversification: Correlations across different such pairs is (presumably) low, so a big

loss from one pair will not coincide with a big loss from another.
▶ However, the risks can turn out to be much larger if these correlations are
mismeasured, or if they change. It is well-known from statistics that correlations are
very hard to measure precisely!

When the prices of two similar assets diverged, and a rational explanation can be given
for why they should converge back to normal, LTCM buys the relatively underpriced asset
and sells the relatively overpriced one.

HKUST, FINA3203, G. Panayotov

 LTCM (cont’d)

They did their arbitrages on a huge scale, with huge leverage (using borrowed funds).

To conserve equity, LTCM financed most of its trades with repos or reverse repos
with six to 12 month maturities. So, LTCM bought a bond and used it as collateral
for a loan, the proceeds from which it used to buy a new bond and used it as
collateral for a new loan, and this could go on and on and on ... (hyper leverage!)

They also did over-the-counter total return swaps, which we will mention later in this
course. At times, they had thousands of open derivatives positions with a total
notional value above $1 trillion!

Their equity was about $5 bn, but their borrowing could exceed $100 bn or more.
Given their reputation, banks were confident lending so much.

LTCM were also secretive, so the market could not easily figure out how much debt
they have – each counterparty saw only its piece of the business and not the total
structure of risks. But this also made difficult the netting of margin payments

HKUST, FINA3203, G. Panayotov

 LTCM (cont’d)

LTCM collapsed, due to:

a series of exogenous macroeconomic shocks that undermined some of their
risk-management assumptions

feedback effects that threatened their creditworthiness, sources of financing and

clearing arrangements.
HKUST, FINA3203, G. Panayotov
 LTCM (cont’d)
LTCM was betting that yield spreads among various bonds in worldwide markets
would narrow; but in 1997-1998 they widened everywhere due to a series of events:
▶ Asian crisis of 1997,
▶ Russian default in 1998 and the contagion that spread to Brazil, Turkey, Venezuela ...
▶ political turmoil in Indonesia, Iraq blocked the U.N. weapon inspections ...
▶ China threatened to devalue the RMB because the Japanese yen’s depreciation
▶ ....

So, LTCM’s spreads started moving against their traders’ expectations, which
increased mark-to-market losses. Volatility increased.

Also, by 1998 many other hedge funds imitated LTCM. When trouble started, all
funds tried simultaneously to reduce their exposure and get out of the same trades
that LTCM had.

It then became clear that LTCM’s portfolio was not really diversified: it was in fact
correlated with market-wide moves, and this was exacerbated by the high correlation
between various hedge fund portfolios.
HKUST, FINA3203, G. Panayotov
 LTCM (cont’d)

As losses accumulated, liquidity evaporated. The few major banks that were still
willing to trade with LTCM knew well its distressed condition, and took advantage

To hide its trades, LTCM had often used different counterparties for each leg of a
possibly complex deal.
▶ Which means that each counterparty saw only one leg, and did not know that in fact it
was well hedged by the other leg (e.g., buy one Treasury bond, and sell a very similar
one).
▶ When LTCM got in trouble, these counterparties assumed the worst and overestimated
LTCM’s true risk.

There were also rumours that LTCM might file for bankruptcy protection in the
Cayman Islands. Counterparties were uncertain of their rights under Cayman law,
which further reduced their willingness to deal with LTCM.

HKUST, FINA3203, G. Panayotov

 Bailout of LTCM

Wall Street feared that LTCM’s failure could start a chain reaction in numerous
markets, causing catastrophic losses throughout the financial system.

After LTCM failed to raise more money on its own, it was running out of options.
▶ On September 23, 1998, Goldman Sachs, AIG, and Berkshire Hathaway offered to buy
out the fund’s partners for $250 million

▶ The offer was stunningly low to LTCM’s partners because at the start of the year their
firm had been worth $4.7 billion.

▶ Warren Buffett gave Meriwether less than one hour to accept the deal.

▶ The deal was rejected by LTCM.

Seeing no options left, the Federal Reserve Bank of New York organized a bailout of
$3.625 bn by the major creditors to avoid a wider collapse in the financial markets.
▶ LTCM’s partners received a 10% stake, worth about $400 million

HKUST, FINA3203, G. Panayotov