Recently, I came to know of a practice by a bank where in they lend silver to their clients. This set off a number of questions in my mind. One of these questions is - does inflation completely explain currency returns in a multi-currency world? And if not, how does one quantify and characterize the non-inflation component of currency returns?
The standard definition of money (and which I believe is still relevant) is that it acts as a measure and store of value. The food we eat, the clothes we wear, the gadgets that we use all have a certain value associated with them, which we can denominate in terms of money. Such denomination makes it easier to buy and sell all goods and services since we have a common reference point.
From this follows the concept of inflation. Because the demand and supply of each good and service varies over a period of time, so does it's price. Hence, at different points of time, the quantity of each good and service that a fixed amount of money can buy is different. This is known as inflation.
In a normal growing economy, prices of goods and services will have a tendency to increase over a period of time. Inflation is positive. Positive inflation represents a decrease in the purchasing power of a currency. If prices keep increasing, one can buy fewer goods and services with the same amount of money.
Now, when one looks upon the amount of money that one has, inflation becomes a measure of its returns. For example, consider that you have one note of 100 rupees. Assume you keep it as it is, stowed safely in your locker. Assume you can buy 4 apples with it, each costing 25 rupees. Now one year passes. Apples now cost 50 rupees. So you can buy only 2 apples with it. Inflation is 100% (because what cost 25 rupees earlier now costs 50 rupees). Then, the return on your currency holdings is exactly equal to -100% (the negative of the inflation)
If there was only one economy in the world and only one currency, inflation and the return on that currency would be the same. Of course, ideally to measure inflation perfectly, one would need to determine the price of each good and service that can be bought or sold. Then one can calculate inflation over one year as the prices of all goods and services now, divided by the prices of all goods and services one year back. Of course, the prices will need to be multiplied by the quantity of each good or service as well. The "basket" should be as broad as possible. So far so good.
Now comes the fun part. When you have multiple economies and currencies, as we do in the world today, what happens? Does the return on one currency still equal the inflation in that currency?
Consider an Indian consumer. His currency is the Indian rupee. Previously, when India was a completely closed economy, the basket of goods and services that he could buy with that currency was restricted to Indian goods and services. Now, however, since India is a relatively more open economy, he can buy certain goods from other parts of the world as well, say China, US, Europe and so on. Thus, to compute inflation now for India, one must look at the global basket of goods and services that Indians can buy and note price rises in each of them. Since these prices will all be denominated in different currencies, they will need to be converted to Indian rupees. Of course, this is much easier said than done, since a number of difficulties are bound to arise.
Now when you think about it, it's not necessary that inflation will completely explain the returns in the rupee. Exchange rates between countries are affected by a number of factors, broadly categorized as trade transactions and capital flows.
Let us consider a simplified example. Assume there are only two countries in the world - the US and India. The USD-INR exchange rate is 50. An Indian consumer can buy a certain basket of goods with his stock of Indian rupees. Assume that the inflation in India is 10% and that in the US is 2% and the Indian consumer spends half of his total money on Indian goods and the other half on US goods. The effective inflation for him would be 6% (taking a weighted average) but only if the exchange rate remains constant. However, the exchange rate will not remain constant because of a multitude of reasons.
The concept of interest rate parity will come in here. In simple terms, interest rate parity tries to account for different rates of inflation in different currencies and says that the exchange rate between two currencies adjusts to account for the difference in inflation rates. So in the above example, if the USD-INR rate one year back was 50 and inflation rates are 10% and 2%, the new rate will adjust to 54 (reflecting the difference of approximately 8%). In such a scenario, inflation will completely explain currency returns.
The problem is that interest parity does NOT always hold. Or rather, we have to be careful in the way we use interest rate parity. Interest rate parity only accounts for the relative changes in purchasing power between two currencies OVER A PERIOD OF TIME. It doesn't account for other factors that cause changes in the CURRENT exchange rate. The exchange rate changes because of reasons other than expected inflation differentials. These include trade and capital flows, economic fundamentals, monetary and fiscal policy, degree of openness of an economy and so on. If this happens, then the return of a particular currency will not be fully explained by inflation. Of course, one would need to do a lot of work and account for a lot of factors before concluding the same.
Now, finally, coming to the link with lending and borrowing.
When one lends money, the interest rate or yield charged perform a function analogous to price. When one looks at the loan as an asset, it has two components that generate return:
- The loan itself i.e. credit as an asset
- The currency in which the loan is denominated
Now the yield (or price) will be a function of supply and demand for both the components i.e. the supply and demand for credit, and the supply and demand for the currency.
The supply and demand for the currency will be influence by a number of factors. The primary supply will get dictated to a large extent by the monetary authority i.e. the central bank, its base interest rates and so on. The demand will be a function of several factors.
Usually, the compensation which is charged by the the lender for lending under the second component (currency), is the expected inflation. However, as we have discussed, the value of the currency over a period of time may not be determined by inflation alone.
Now, if it is true that inflation does not completely explain returns in that currency, then when a lender lends to a borrower, in addition to expected inflation, there should be an adjustment in the yield reflecting that portion of expected returns on the currency which is not explained by inflation.
Also, the non-inflation portion of expected returns can be both positive or negative. Thus, un-hedged lenders in a currency should usually demand a risk premium for a currency that is more volatile against other currencies, to compensate for this two way risk.
A lot of work on this area remains to be done. Comments & feedback most welcome.
The standard definition of money (and which I believe is still relevant) is that it acts as a measure and store of value. The food we eat, the clothes we wear, the gadgets that we use all have a certain value associated with them, which we can denominate in terms of money. Such denomination makes it easier to buy and sell all goods and services since we have a common reference point.
From this follows the concept of inflation. Because the demand and supply of each good and service varies over a period of time, so does it's price. Hence, at different points of time, the quantity of each good and service that a fixed amount of money can buy is different. This is known as inflation.
In a normal growing economy, prices of goods and services will have a tendency to increase over a period of time. Inflation is positive. Positive inflation represents a decrease in the purchasing power of a currency. If prices keep increasing, one can buy fewer goods and services with the same amount of money.
Now, when one looks upon the amount of money that one has, inflation becomes a measure of its returns. For example, consider that you have one note of 100 rupees. Assume you keep it as it is, stowed safely in your locker. Assume you can buy 4 apples with it, each costing 25 rupees. Now one year passes. Apples now cost 50 rupees. So you can buy only 2 apples with it. Inflation is 100% (because what cost 25 rupees earlier now costs 50 rupees). Then, the return on your currency holdings is exactly equal to -100% (the negative of the inflation)
If there was only one economy in the world and only one currency, inflation and the return on that currency would be the same. Of course, ideally to measure inflation perfectly, one would need to determine the price of each good and service that can be bought or sold. Then one can calculate inflation over one year as the prices of all goods and services now, divided by the prices of all goods and services one year back. Of course, the prices will need to be multiplied by the quantity of each good or service as well. The "basket" should be as broad as possible. So far so good.
Now comes the fun part. When you have multiple economies and currencies, as we do in the world today, what happens? Does the return on one currency still equal the inflation in that currency?
Consider an Indian consumer. His currency is the Indian rupee. Previously, when India was a completely closed economy, the basket of goods and services that he could buy with that currency was restricted to Indian goods and services. Now, however, since India is a relatively more open economy, he can buy certain goods from other parts of the world as well, say China, US, Europe and so on. Thus, to compute inflation now for India, one must look at the global basket of goods and services that Indians can buy and note price rises in each of them. Since these prices will all be denominated in different currencies, they will need to be converted to Indian rupees. Of course, this is much easier said than done, since a number of difficulties are bound to arise.
Now when you think about it, it's not necessary that inflation will completely explain the returns in the rupee. Exchange rates between countries are affected by a number of factors, broadly categorized as trade transactions and capital flows.
Let us consider a simplified example. Assume there are only two countries in the world - the US and India. The USD-INR exchange rate is 50. An Indian consumer can buy a certain basket of goods with his stock of Indian rupees. Assume that the inflation in India is 10% and that in the US is 2% and the Indian consumer spends half of his total money on Indian goods and the other half on US goods. The effective inflation for him would be 6% (taking a weighted average) but only if the exchange rate remains constant. However, the exchange rate will not remain constant because of a multitude of reasons.
The concept of interest rate parity will come in here. In simple terms, interest rate parity tries to account for different rates of inflation in different currencies and says that the exchange rate between two currencies adjusts to account for the difference in inflation rates. So in the above example, if the USD-INR rate one year back was 50 and inflation rates are 10% and 2%, the new rate will adjust to 54 (reflecting the difference of approximately 8%). In such a scenario, inflation will completely explain currency returns.
The problem is that interest parity does NOT always hold. Or rather, we have to be careful in the way we use interest rate parity. Interest rate parity only accounts for the relative changes in purchasing power between two currencies OVER A PERIOD OF TIME. It doesn't account for other factors that cause changes in the CURRENT exchange rate. The exchange rate changes because of reasons other than expected inflation differentials. These include trade and capital flows, economic fundamentals, monetary and fiscal policy, degree of openness of an economy and so on. If this happens, then the return of a particular currency will not be fully explained by inflation. Of course, one would need to do a lot of work and account for a lot of factors before concluding the same.
Now, finally, coming to the link with lending and borrowing.
When one lends money, the interest rate or yield charged perform a function analogous to price. When one looks at the loan as an asset, it has two components that generate return:
- The loan itself i.e. credit as an asset
- The currency in which the loan is denominated
Now the yield (or price) will be a function of supply and demand for both the components i.e. the supply and demand for credit, and the supply and demand for the currency.
The supply and demand for the currency will be influence by a number of factors. The primary supply will get dictated to a large extent by the monetary authority i.e. the central bank, its base interest rates and so on. The demand will be a function of several factors.
Usually, the compensation which is charged by the the lender for lending under the second component (currency), is the expected inflation. However, as we have discussed, the value of the currency over a period of time may not be determined by inflation alone.
Now, if it is true that inflation does not completely explain returns in that currency, then when a lender lends to a borrower, in addition to expected inflation, there should be an adjustment in the yield reflecting that portion of expected returns on the currency which is not explained by inflation.
Also, the non-inflation portion of expected returns can be both positive or negative. Thus, un-hedged lenders in a currency should usually demand a risk premium for a currency that is more volatile against other currencies, to compensate for this two way risk.
A lot of work on this area remains to be done. Comments & feedback most welcome.
Posts
