Covered Interest Arbitrage

Covered Interest Arbitrage

Self-paced Tutorial

Online tutorials and registration for mini classes at: http://tradingroom.bentley.edu

2006© Hughey Center for Financial Services. All Rights Reserved.

1

Table of Contents:

Content

Page

PART I - Objectives and Background

3

PART II - Country selection

4

PART III - Data Collection

6

Currency Background

7

Ringgit Spot Rate

7

Ringgit Forward

8

Ringgit Deposit

8

US Deposit

9

Downloading the Data

9

PART IV - Model Construction

10

2

ARBITRAGE in the INTERNATIONAL FINANCIAL MARKETS COVERED INTEREST ARBITRAGE Self-paced Tutorial PART I – Objective and Background. Objective: The purpose of this self-paced tutorial is to guide you through the analysis and valuation of arbitrage opportunities that routinely appear in the international financial markets. Specifically, in the next few pages, we will analyze both exchange rates and short-term interest rate relationships between the US Dollar (USD) and the Malaysian Ringitt (MYR). Background: o We will build a dynamic model charged with identifying Covered Interest Arbitrage opportunities. Or risk-free profit-generating transactions that leverage on the use of spot and forward exchange quotes along with short-term forward deposit rates. o In order to build the model we will have to collect historical information for the following classes of financial instruments: a) b) c) d)

Foreign Currency: Spot Exchange Rate Foreign Currency: Three-month Forward Quote Foreign Currency: Three-month Deposit Rate US Dollar: Three-month Deposit Rate

We will make use of Reuters Kobra 3000 to gather the ticker symbols for each of the above instruments, then Reuters Power Plus to download historical information for each of these pieces, for a specified time period, into a spreadsheet o It is noteworthy to mention that arbitrage opportunities can be uncovered when comparing a very efficient market, like the US, to a lesser efficient one, like Malaysia. It’s the combination of that two that creates room for arbitrage. Had we selected two very liquid and efficient markets, i.e. the US and the UK, we would have not found much in terms of risk-free profits and Covered Interest Arbitrage opportunities. o The fact that a lesser efficient market’s currency is pegged to the other is really not an obstacle in seeking out arbitrage opportunities. If anything, it might constitute an advantage (we will expand on this concept in the next section). o One can assume a multitude of perspectives in analyzing the relationships that exist in creating the model. Our basic assumption will be to consider a private investor dealing with a financial institution when participating in the process. This implies that we will always be on the worst side of the deal (in terms of bid/ask) when exchanging currency and/or making a money market investment.

3

PART II – Country Selection As mentioned above, the country selection process is of outmost importance. The key for seeking out these risk-free profits is to look for an exotic market that provides a certain level of recurrent market inefficiency. Let’s assume we want to determine if any CIA opportunities exist between the Malaysian Ringgit and the US Dollar. As a first step, we can look at the currency’s graph over time on Bloomberg. The screen shot below plots monthly market prices for the Malaysian Ringgit from January 1, 1993 through September 30, 2003.

Interestingly enough, we can see that something happened in the fourth quarter of 1998 which caused the exchange rate to go flat ever since. What do you think may have caused such behavior in the currency value? In order to answer that question, we need to make sure that what appears to be a flat exchange rate is actually true. In order to do this, we can zoom in on a more narrow range of dates and look at the pattern. Specifically, let’s go back to the Bloomberg terminal and view a graph of the weekly market prices of the Ringgit from January 31, 2001 to September 26, 2003. The screen shot below illustrates this particular section of the graph above.

4

Is this surprising? Actually not. What happened here is that what looked to be a flat line in the previous screen, turns out to be a very precise trading range in which the Ringgit fluctuates. In order to understand what happened we have to do a little more research, on Bloomberg again. Looking back at the news releases being sent across the wires on and around September 13, 1998, you will notice an interesting on dates 9/13/1998. The caption below is only the first of the six-page article in question. One week back, the Malaysian Government decided to peg (to fix) the Ringgit to the US Dollar at 3.8 Ringgit per 1USD. This release adds that the Malaysian Government is implementing a 2% buffer zone around the fixed rate of 3.8 with a high and low range. As a result of establishing this trading corridor, the exchange rate for the Ringgit exhibits is commonly described as a “snake” pattern. It fluctuates up and down within its corridor. Digressing for a moment from the Arbitrage setting, one can appreciate that knowing exactly how high and how low the rate will go can be leveraged to make buy and sell decisions. In other words, if we see that the MYR exchange rate is at the lower end, we can Sell the Ringgit at its strong point and Buy it back when it reaches the higher end of the spectrum, or when it’s going to be worth less Dollars.

5

PART III – Data Collection Before we actually create a dynamic model that can seek out arbitrage opportunities for us to exploit, we need to take a few more steps and collect the necessary data for the model. Specifically, we will need to retrieve Daily Bid and Ask quotes for the following 4 items for a threemonth period, in this case will chose April 1, 2002 though October 10, 2002. This constitutes 90 data points, or three months worth of actual data: 1) 2) 3) 4) 5)

Currency Background Page Malaysian Ringgit: Spot Exchange Rate Malaysian Ringgit: Three-month Forward Quote Malaysian Ringgit: Three-month Deposit Rate US Dollar: Three-month Deposit Rate

We will use Reuters 3000 Xtra to get the ticker symbols for each one of these instruments. By launching Reuters PowePlus we’ll then download the historical information into a spreadsheet. Launch Reuters Kobra 3000 Xtra by doudle-clicking on the icon on your desktops. Go into the REUTERS Speed Guide and double click between the brackets to get to the information.

6

1) Currency Background Page The Currency background Page is a critical page for the purpose of this model, it allows us to see what the spot ticker is for the Malaysian Ringgit, whether its quote is expressed in direct or indirect terms against the US Dollar, and the decimal precision point we should use.

You have to follow the path below to get to the screen shot presented here: <MONEY>, <MONEY / BKGDINFO1>, <MONEY/BKGDINFO2>, <MYR/BKGDINFO>. 2) Malaysian Ringgit: Spot Exchange Rate The spot rate for the MYR= can be found following the path below. This rate is the one at which you could exchange your money for right now. The path is: <MONEY>, <SPOT / 1>, , <MYR=>.

7

3) Malaysian Ringgit: Three-month Forward Quote The three-month forward quote represents what the investment community thinks the exchange rate will be three months from now. By purchasing a forward contract you lock in the rate. Three months from today, you will be able to exchange your money at that rate regardless of market conditions. The path below illustrates how to get to this quote window in Reuters: <MONEY>, , <MYRF= >, <MYR3M= >.

4) Malaysian Ringgit: Three-month Deposit Rate The three-month deposit rate represents the interest rate you would earn if you invested for a threemonth period in a Malaysian money-market account. It also represents the rate you would be charged for borrowing money on a three-month basis in Malaysia. Follow the path below: <MONEY>, , <MYRF= >, <MYR3MD= >.

8

5) US Dollar: Three-month Deposit Rate The three-month deposit rate represents the interest rate you would earn if you invested for a threemonth period in a US money-market account. It also represents the rate you would be charged for borrowing money on a three-month basis in the US. Follow the path below: <MONEY>, , , , .

6) Downloading the Data Once you have collected the four ticker symbols for your set of countries, you will need to download historical daily data for each, individually. Follow the steps below: You should already have Reuters Kobra 3000 Xtra open, don’t close it, minimize it. Next Launch Reuters PowerPlus Pro , an Excel-like icon on your desktop. From the main menu at the top select: Reuters > Real-time Data > History. This will launch Real Time History Subject Selection. There are three (3) steps in this process: Step 1: Enter the ticker symbol in the “Code” field, ALL IN CAPS,.and click ADD. NEXT Step 2: Click the “Applicable Fields” folder in the “Fields by Category” box. The Date category is already selected for you. Double click the options “Bid” and “Ask” that appear under the field column in the box below and they should appear under the “Selected Fields” box. NEXT Step 2a: Select the time period for your historical data list. For the purpose of this model, you should select DAILY, this will allow you to download the daily closing prices only. For the purpose of this assignment you will want to enter “90” into the field “Data Points.” In the field “End Date” you should select the end date as you like. NEXT Note: the Reuters database is equipped with two years of daily data, five years of weekly data, and 10 years of monthly data.

9

Step 3: Click “Finish” and the data should appear on your spreadsheet. The data you just collected should be arranged in an Excel spreadsheet to resemble the screen shot below (with the exception of the actual dates and numbers):

PART IV – Model Construction Now that we have collected the data we just need to re-arrange it to reflect the needs of our model and the stated objectives. Seek out Covered Interest Arbitrage opportunities. 1) The first step to take is to translate the Forward Quote that we received from Reuters into a Forward Outright Quote. Meaning that Reuters tells us how many points (i.e. 135, 150)we have to add to the current spot rate to get the forward. The formula to obtain a FWD Outright is: [MYR Bid + (MYR3M Bid/10,000)] = MYR3M Bid outright Æ 3.7997+(135/10000) = 3.8130. Had the two Fwd quotes been negative, we would have subtracted them from the MYR spot. The only trick here is to remember that the Bid is always lower than the Ask. Our Forward outright Bid would therefore be the MYR Bid – MYR3M Bid/10,000, and vice versa for the Forward Outright Ask. [MYR Bid - (MYR3M Bid/10,000)] = MYR3M Bid outright. You need to create two new columns where you enter the Forward Outright Quote, Bid and Ask, for each day. Once you’re done, the spreadsheet should now look like this:

10

2) The second step is to transform all of the above quotes and express them in US Dollar terms. As of right now, they are all expressed in Ringgit terms (USD/MYR), we want MYR/USD. The way to do this is to divide both Bid and Ask for the Spot quote and the Forward Outright quote by 1, also called reciprocal. The only trick here is to keep in mind that the Bid has to be always lower than the Ask. The MYR/$ Bid is equal to 1/($/MYR) Ask, and the MYR/$ Ask = 1/($/MYR) Bid. In other words you need to divide 1 by the Ask to get the Bid and vice versa for the Ask. Same goes for the forwards. As for the interest rates, we need to express them in three-month terms. Interest rates are always quoted on an annual basis but we are dealing with three-month arbitrage here, so we need to divide by four, and add 1. The only reason why we add 1 is to make things easier later on when we use the formulas. The screenshot below captures what the model should look like after the transformations.

3) The third step in this model consists of establishing the criteria for identifying arbitrage opportunities. The idea here is to borrow in the country that has a lower interest rate than what you can earn by investing the funds, exchanged, in the other country. The two formulas below detail the sequence of steps you have to follow to identify and exploit any arbitrage opportunities. There are two different formulas to reflect the two different strategies. Formula 1: We borrow in the US and invest in Malaysia. Formula 2: We borrow in Malaysia and invest in the US. Formula 1 Step-by-Step (10/08/03) step 1 borrow USD ($1M) will repay USD ($1M * USD3MD ask) step 2 conv in MYR @spot [$1M / (MYR/$) Ask] step 3 Invest in MYR (3,799,500 * MYR3MD Bid) step 4 buy FWD USD [3,827,331 * (MYR/$) Bid] step 5 payback USD loan (1,003,232-1,002,925)

$1,000,000 $1,002,925 3,799,500 3,827,331 $1,003,232 $307

profit

11

Step 1: What we have done here is borrow $1M in the US and record that we will have to repay the principal plus the interest in three months. We use the Ask, remember the “rip-off rule”, we are always on the worst side of the deal. We would like to pay less interest (Bid) but will pay more (Ask). Step 2: Convert the USD into MYR at the spot rate today. Again, we would like to get more Ringgit (Bid) but we’ll have to convert at the Ask. Step 3: Once we have the foreign currency on hand we can invest it in Malaysia and earn a threemonth deposit rate (Bid). Step 4: At this point we need to buy enough three-month forward contracts to lock in our profit. We do that by converting (multiplying) the proceeds from the Malaysian investment at the three-month forward rate (Bid). Step 5: At this point we are back in USD and we just have to repay the initial loan. Anything left is a profit. Formula 2 Step-by-Step (10/08/03) Step 1 borrow $1M worth of Ringgit [$1M / ($/MYR) Ask] 3,799,500 will repay MYR (3,828,281 * MYR3MD Ask) 3,828,281 Step 2 conv. in USD @ spot [3,799,500 * ($/MYR) Bid] 999,737 Step 3 Invest in US (999,737 * USD3MD Bid) 1,002,461 Step 4 buy FWD MYR [3,828,281 * ($/MYR) Ask] 1,004,008 Step 5 payback MYR loan ( 1,002,461-1,004,008) (1,547)

loss

Formula 2 follows the same logic as Formula 1, except that this time the countries in which we borrow and invest are reversed. As you can see if we use the same data in both formulas we will get two different results. This is generally true, you will find arbitrage opportunities in one country or the other, not both. A good way to implement these formulas into the model is to create two columns next to the quotes and rates where you compute each of the two formulas for each row of data. This way you can immediately see if, and where, arbitrage exists.

12

You can see from the image that May 30 and July 9 presented arbitrage opportunities. Note that row 5 has a different name; cell A5 is called Today RT and doesn’t actually have a date. What we’ve done here is link cells A5 through K5 to the real-time Reuters feed. By doing this, we enable the spreadsheet to represent the most current information at any point in time. And because the FX doesn’t close, we can get this information real-time around the clock. How to do this is quite simple: if you recall PART III, points 2 through 5, we used Reuters Kobra 3000 Xtra to retrieve the ticker symbols for the four instruments. Let’s start with the spot rate, just go back to that window the drag and drop the Bid and the Ask in cells B5 and C5. You literally place the cursor over the Bid quote, click, drag it into B5 and release.

Click on the Bid quote (3.7995) and drag it into cell B5

You can do the same for the Ask, and you drag and drop any of the other information fields contained in the window. If your model is already built, these changes will flow through the spreadsheet automatically.

13