The Chinese currency for Macro econ Class

chapter10_nm2.pdf

Home >Business & Finance homework help >Economics homework help >The Chinese currency for Macro econ Class

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Chapter 10: Exchange Rates, Interest Rates, and the Foreign Exchange Market

The exchange rate is the key relative price for an economy open to international trade and finance. The price of Japanese made automobiles imported into the United States depends on how many dollars it takes to buy 100 yen, i.e., the dollar/yen exchange rate. Whether U.S firms can profitably sell heavy earth- moving equipment like bulldozers to the rest of the world is partly determined by the strength of the dollar against the currencies of the countries the US exports those machines to. And whether one decides to purchase in U.S. stocks and bonds, as opposed to say German securities, has a great deal to do with how one expects the dollar to move against the euro over time.

In this chapter, we’ll take the first steps to understanding what that the exchange rate is, and how its value is determined. We start this process by first defining terms, second describing the marketplace for currencies, and third discussing the linkage between exchange rates and other asset prices that are linked to exchange rates.

10.1 Exchange Rates and Currency Trading

In the real world, there are many exchange rates; after all the term means the rate of exchange between two items. Most commonly, the exchange rate pertains to the rate of exchange of two currencies. Other meanings include the rate of exchange of real goods, or between a currency and bundle of other currencies. We’ll first focus on the first definition, and relate it to trading in currencies. At the end of the chapter, we’ll discuss exchange rates between real goods and services, and how those relate to exchange rates between currencies.

In the most familiar case, the exchange rate is the number of home currency units required to purchase one unit of foreign currency. In other words, the exchange rate is the price of foreign currency. In Chart 10.1, three exchange rates against the US dollar are shown. In each case, the values refer to the number of US dollar necessary to purchase a single unit of the foreign currency at a given time (sometimes known as a spot exchange rate). Notice that the Deutsche mark disappears in 1999 because it is subsumed into a new currency, the euro.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

0.0

0.5

1.0

1.5

2.0

2.5

3.0

60 65 70 75 80 85 90 95 00 05 10 15

XGR XEU XUK

USD/DEM USD/EUR

USD/GBP

Exchange rates against US dollar

Chart 10.1: Exchange rates for Deutsche mark (blue), British pound (green) and euro (red) against USD. Higher values denote a weaker dollar against the foreign currency (a depreciation).

This exchange rate can best be thought of as a relative price of assets, in this case two currencies. Like many asset prices, the exchange rate reflects the valuation of these assets, which in turn depends on a variety of factors, both those in the present, and those expected in the future. Since those expectations change a lot over time, so too does the exchange rate – as shown by the wide swings in each of the series in the graph over the past 40 years. The idea of expectations and exchange rates is pursued further in Chapter 15.

For now, we take a first try at explaining how are these prices of foreign exchange are determined. At the basic level, the answer is simple – prices are determined in a marketplace. The marketplace, however, is very special. First, it’s extremely large: daily turnover in April 2013 was USD 5.3 trillion per day. It’s global in nature, with trading taking place almost around the clock. And third it’s dominated by a few players: approximately three-quarters of global foreign exchange trading is accounted for by the top ten banks, as shown in Table 10.1.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Table 10.1: Share of global foreign exchange in 2015. Source: Euromoney.

The market is an “over the counter” market; hence there is no centralized exchange. Rather, trading takes place between a set of actors – dealers, non-dealer financial institutions, corporations and governments. The latter are “end-users” – that is they demand currencies for use in their activities. In contrast, dealers trade on behalf of their clients, or they trade in order to make profits for their own financial institution.

In recent years, the development of new computer technologies has lowered transactions costs in foreign exchange markets. As a consequence, these actors have been joined by retail investors and by computer- automated traders known as “algorithmic” traders.

The foreign exchange market encompasses trades in different types of assets, as shown in Table 10.2. Of the USD 5.3 trillion in turnover in 2013, a little less than 40% is accounted for by spot transactions – a single transaction involving an exchange of currencies – and the remainder by derivative transactions. These derivative transactions include outright forwards, foreign exchange swaps, and currency swaps. A full 42% is accounted for by foreign exchange swaps – an exchange of currencies for a predetermined amount of time. These derivatives are further explained in Box 10.1

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Table 10.2: Foreign exchange turnover. Source: BIS (2016).

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Box 10.1: Foreign Exchange Transactions

Source: BIS (2010).

Most of the trading volume is between dealers, while the next most with financial institutions. Non- financial firms followed behind in importance. This pattern highlights the fact that currency demand is mostly underpinned, not by trade in goods and services, but rather by the need to buy and sell financial assets. This demand is in turn derived in part from speculative motives – the desire to acquire assets in anticipation of asset price increases – and in part from the desire to move financial capital across borders.

The pattern of trading in currencies illustrates the relative importance of the motivations. As shown in Table 10.3, the US dollar, the euro, Japanese yen, and British pound lead in terms of currency trading turnover. The Australian dollar recently supplanted the Swiss franc as the fifth place holder. The US dollar is the outsized leader; and yet this outcome cannot be because the US is the biggest exporter or importer of goods and services. Rather, it’s the importance of US financial assets – Treasury bills and bonds and stocks – that drives the dollar’s importance.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Table 10.3: Foreign exchange turnover. Source: BIS (2016).

As noted before, the Australian dollar has recently climbed in importance, reflecting its economic role in a part of the world that has grown in importance. Interestingly, despite their economic mass, the currencies of the BRICs – Brazil, Russia, India and China – account for only 5.9% of total turnover (of 200%). Despite its substantially larger economy, China’s renminbi accounted for 2.2%.

These turnover figures do not fully reflect the central role of the dollar. Not only is dollar turnover high, the dollar is involved in 87% of all currency trades. This is shown in Table 10.4.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Table 10.4: Global foreign exchange market turnover by currency pair. Source: BIS (2016).

The US dollar/euro pair alone accounts for 24% of total turnover. The dollar/yen pair is next in importance. Notice that the high volume pairs all involve the US dollar, highlighting the dollar’s role as the key global currency. The euro/yen and euro/pound pairs account for a mere 4.7% of total trading turnover.

When the objective is to exchange, for instance, Swedish krona for Thai baht, it turns out to be less expensive to exchange krona for dollars, and then dollars for baht. That’s because the market for dollars is so active that the transactions costs are much smaller.

The figures in Table 10.3 also prove the durability of these trading patterns. The dollar/euro pair remained the most important trading pair throughout the decade. Before 2001, the dollar/deutsche mark held that position consistently for the preceding two decades. The dollar/yen pair has been the second most important pair over the period that the surveys have been conducted (about 25 years).

Why is the dollar so dominant in all these foreign exchange transactions – far disproportionate to its size as measured by GDP? This is a complicated question, but it comes down to the question of what makes an “international” currency. This is addressed in Box 10.1.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Box 10.2: Why Is the Dollar the World’s International Currency?

An international currency is one that fulfills the roles of money – unit of account, medium of exchange, and store of value – for both private actors and governments. Private actors use certain currencies primarily because of patterns in trade and finance, but also because of the ease with which those currencies can be traded. Governments also hold foreign currencies.

Box Table 10.1: Functions of an international currency. Source: Kenen (1983). The dollar has been the dominant currency in the holdings of central banks since the end of World War II, and as shown in Box Chart 10.1, has retained that lead even as America’s share of world GDP has shrunk to about a quarter. The only other currency that really registers on the graph is the euro. The Japanese yen and the British pound vie for third place. One notable absence is the Chinese currency, the yuan.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

USD

DEM

EUR

GBP JPY

Esti mated USD

Esti mated EUR

CHF

Reserve shares

Box Chart 10.1: Currency shares in central bank foreign exchange holdings. Source: IMF, Annual Reports, Chinn and Frankel (2008), and IMF, Currency Composition of Official Exchange Reserves (COFER), June 30, 2014. Data for USD and EUR since 2000 are estimated. Statistical analyses suggest that the determinants of the composition of reserves depends on the pattern of output and trade, the development of financial markets, the confidence in the currency’s value, and the presence of “network externalities”.

On the first point, the currency of a country that has a large share in international output, trade and finance has immediate advantage over other contenders in terms of becoming a reserve currency, as well as an international currency. The U.S. economy remains the world's largest in terms of output and trade, so it’s no surprise it’s the world’s reserve currency. But the euro zone should be a close (rather than far) number two, given its relative size. Furthermore, Japan barely registered as a reserve currency even before the advent of the euro, when it was the second largest economy.

To attain international currency status, capital and money markets in the home country must be not only open and free of controls, but also deep and well- developed. Nobody wants to hold lots of a currency that is difficult to trade. Hence, the large financial marketplaces of New York and London clearly benefit the dollar and pound relative to the euro, as Frankfurt is still less well-developed. Tokyo and Frankfurt financial markets have changed a lot over the last two decades. But they still lag far behind New York and London as financial centers. Meanwhile, Singapore and Hong Kong have gained.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

In addition, confidence is also key. The more it is believed a currency will retain its value, the more likely it will be held as a reserve currency. With the decline in inflation in the United States since the 1980’s, the dollar’s share has rebounded from the lows recorded in the 1970s.

Finally, the dollar benefits because everybody is already using it, by accident of the US being the dominant economic force right after World War II. Once started, it is hard to switch, unless the dollar becomes sufficiently unattractive on the three other counts, just recounted.

Chinn and Frankel (2007) conducted a statistical analysis of the determinants of the composition of central bank reserve holdings and found that financial development was the key determinant (along with inertia). This finding suggests that -- even with China’s rapid economic growth – as long as financial system remains underdeveloped and largely closed to the rest of the world, the Chinese yuan will not become a major reserve currency.1

Just as trading in currencies is concentrated, so too is the geographic location of foreign exchange trading. However, the geographic distribution does not correspond to what one might expect. By far, as shown in Chart 10.2, the largest center for trading foreign exchange is in the United Kingdom, despite the fact that the pound sterling is not the most traded currency. Related to this point, London – not Frankfurt – is the center for trading euros. This observation demonstrates the fact that trade in a currency need not be concentrated in the country of issue.

Next up is the United States, with Japan a distant third. Trading in the remaining locations – France, Germany, Hong Kong, Singapore, Australia and Switzerland – ranges between 3-4%. In order to reduce clutter, locations have been aggregated.

1 For more discussion of the prospects of the renminbi as a major reserve currency, see Chinn (2015).

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

1 2 3 4 5 6 7

Geographical share of foreign exchange trading, %

HK, Singapore, Australia Japan

France, Germany

Switzerland

Chart 10.2: Source: BIS (2013, 2016).

10.2 Supply, Demand and Exchange Rates under Fixed and Floating Regimes

Thus far, we have discussed the institutional features of the markets wherein the various actors – traders, corporations, central banks – buy and sell currencies. However, we have not explained how these forces interact to determine the price of a particular foreign currency.

It turns out that we can use the standard supply and demand framework as a means of thinking about how the exchange rate – the price of foreign currency – is determined. For the moment, assume the proceeds from exports of goods and sales of assets constitute the supply of foreign currency. Spending on imports and purchases of assets are the sources of demand for foreign exchange. Then consider a diagram where the price of the foreign currency (S) is on the vertical axis, and the quantity of foreign currency on the horizontal.2 In the example shown in Figure 10.1 below, the market for euros is examined, from the perspective of a US resident.

The demand for euros is essentially derived from underlying demand for goods and assets denominated in euros. Similarly, the supply of euros is derived from the underlying demand for supply of goods and assets that euro residents purchase. The demand curve is downward sloping because as euros become cheaper, the quantity demanded rises. The supply curve is upward sloping because as euros become more expensive, the quantity supplied increases.

2 “S” is short for the spot exchange rate, the price for a current transaction. This is to distinguish it from the price in a contract to transact in the future, what is called “forward exchange rate”, or “F” for short.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Figure 10.1: Supply and demand for euros

The equilibrium exchange rate – when the market is allowed to determine the price as it is under a floating rate regime – is S0. At this point, the quantity supplied equals the quantity demanded.

When supply or demand changes, then the exchange rates responds. Consider for instance what happens if the demand for euros increases – perhaps because the demand for European goods exogenously increases. Then the demand curve shifts out, and the exchange rate rises from S0 to S1 (Figure 10.2).

Figure 10.2: An increase in demand for euros

Supply

Demand

S{$/€}

Demand' S0

€

S{$/€}

€

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Notice that when the exchange rate rises, this is the same as the price of foreign exchange rises – that is foreign currency becomes more expensive. Viewed from the other side, the home country currency (the dollar in this case) loses value. Hence, the rise in E represents a dollar depreciation.

Next we consider what happens if the supply of foreign currency rises, as shown in Figure 10.3. This could happen if foreign demand for US bulldozers increases.

Figure 10.3: An increase in supply of euros

The equilibrium exchange rate falls from S0 to S1. This represents a decrease in the price of euros in terms of dollars, so the dollar is becoming more valuable. Hence, this is an appreciation of the dollar.

In the two preceding examples, we’ve analyzed the repercussions of shifts in supply and demand in a system where the exchange rate is free to adjust to equilibrate the quantities supplied and demanded. This type of regime is called a freely floating exchange rate regime.

In many cases, the exchange rate is not allowed to freely move; sometimes the exchange rate is fixed – or pegged – to a specific value against another currency. In fact, the extreme stability of the dollar/Deutsche mark and the dollar/pound exchange rates in the late 1960’s to early 1970’s shown in Figure 10.1 hinted at this possibility. When the exchange rate is fixed, somebody has to step in to maintain that price. In foreign exchange markets, that somebody is the central bank. The central bank’s role is easiest to explain by resorting to the same type of supply-demand graph used above. Now, instead we’ll take the perspective from a Chinese resident (for many years up until July 2005, China effectively pegged their currency, the Chinese yuan, to the US dollar).

Figure 10.4 depicts the market for US dollars (which is the foreign currency in this example). The exchange rate is set to SPeg , expressed in the number of Chinese yuan (CNY) necessary to purchase one single US dollar (so it’s drawn from the perspective of a Chinese resident).

Supply

Demand

S{$/€}

Supply'

€

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Figure 10.4: Chinese supply and demand for dollars under a fixed exchange rate regime.

In this case, the exchange rate peg is set above the equilibrium exchange rate such that there is excess supply of dollars in the foreign exchange market. In the absence of any other forces, the exchange rate will gravitate to S1, that is the Chinese yuan will appreciate.

However, under a fixed exchange rate regime, the central bank commits itself to maintaining the currency at the official exchange rate. Hence, the central bank, in this case the People’s Bank of China, would have to absorb the excess supply of dollars by purchasing those dollars in the foreign exchange market. In essence, the central bank creates the amount of dollar demand necessary to set the exchange rate at the pegged level.

Figure 10.4 depicts a situation where the peg is set above the equilibrium rate. We can consider the opposite case, where the peg is set below the equilibrium rate. Then there is excess demand for dollars, which the central bank must satisfy by selling dollars.

In sum, if policymakers aim to keep the exchange rate at a depreciated level relative to the equilibrium determined by private supply and demand, then the central bank must on net purchase foreign currency; if it seeks to keep the exchange rate at an appreciated level, then the central bank must sell foreign currency.

The preceding discussion has set forth a sharp distinction between floating and fixed exchange rate regimes. In practice, countries are seldom purists regarding the exchange rate regime they pursue. Rather, countries – particularly emerging market and less developed countries -- will often manage exchange rates, intervening to smooth movements in rates.

Why might a country’s policymakers seek to fix the exchange rate? One reason is that eliminating variations in the exchange rate makes planning by exporters, importers and those wishing to borrow and lend in foreign currencies, a lot easier. For instance, if a Chinese exporter was considering to build a new factory to sell steel to the United States, but the factory would take a couple years to complete, the

Supply

Excess supply $

S{¥/$}

Demand

SPeg

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

decision whether to proceed or not would be much easier if one knew what the yuan/dollar exchange rate would be in two years.

On the other hand, keeping the exchange rate fixed imposes, if not certain costs, at least certain trade-offs in terms of what the policymakers can do with respect to monetary policy and the free flow of capital across the border. Those trade-offs are discussed in Chapter 13.

10.3 Exchange Rate Regimes in the Real World

Each year, countries report to the International Monetary Fund the exchange rate arrangements they claim to follow. The IMF reports in the Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER) these de jure policies for 2012. The tabulation is reported in Table 10.5. At the top are the most rigid, or fixed, exchange rate regimes. As one reads down the table, the exchange rate regimes become more flexible. Reading across the table, one can identify the currency which is used as the anchor – most currencies are linked to the dollar or the euro, with some others to a basket of currencies (“composite”).

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Table 10.5: De Facto Classification of Exchange Rate Arrangements and Monetary Policy Frameworks, April 30, 2012. Source: Table 1, from IMF, Annual Report on Exchange Rate Arrangements and Exchange Restrictions, 2012.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

It is interesting that there are quite a few countries that assert that they peg the exchange rate (43), as well free floating (30). But even more numerous are those that claim to follow an intermediate regime (56)! Another 35 claim to be floating, which could also be interpreted as an intermediate regime.

The AREAER categorizations are a useful summary of the type of exchange rate regimes in place. However, we would like to also know the degree to which the exchange rate is allowed to move, once a country is not on a hard fixed exchange rate regime.

Carmen Reinhart and Kenneth Rogoff have taken up that challenge. They classified exchange rate regimes by their degree of flexibility. Rather than categorizing regimes by their stated policies (e.g., de jure), but rather by how exchange rates actually behave. They used the following criteria to classify exchange rate regimes into 15 categories, with the lowest value indicating the highest degree of exchange rate rigidity, and the highest value indicating the highest degree of exchange rate flexibility. The distribution of exchange rate regimes for 2000 is shown in Chart 10.3.

Chart 10.3: Frequency histogram of de facto exchange rate regimes. Source: Reinhart and Rogoff (2004).

Category 1 pertains to countries with no separate legal tender – that is they are part of a currency union, such as the euro. Category 15 – “free falling” – describes currencies that are losing value as in a crisis situation. Category 2 corresponds to what was described as a fixed exchange rate regime in Section 10.3, while Category 14 corresponds to floating exchange rate regimes.3

3 The full description is as follows: No separate legal tender (1), Pre announced peg or currency board arrangement

0 .0

5 .1

.1 5

.2 F

ra ct

io n

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Exchange rate regime used in Reinhart, Carmen and Kenneth Rogoff,(2002)

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Chart 10.3 reflects the fact that few countries manage their currencies in such a way as to fit our characterization of prototypical exchange rate regimes. Most countries allow some flexibility, limiting day-to-day changes in currency value. The countries that allow their currencies to float still manage their rates either by use of occasional foreign intervention (buying and selling currencies in the foreign exchange market), or by changing interest rates in order to influence currency values.

The correlation between the de jure and de facto measures is surprisingly low. Even when the Reinhart and Rogoff categories are collapsed into 3 aggregate categories, the correlation is only 33%, and the percentage of coincidences is only 55%.4

10.4 The Relationship between Exchange Rates and Interest Rates

Why do households and firms trade currencies? Part of the reason is that they need foreign currency for purposes related to their activities that cross border – for instance buying products from another country. That was the key motivation for the shifts in the supply and demand curves in Section 10.2. However, that’s not the only motivation. Firms in particular often have funds that they wish to get the highest rate of return. How can we analyze the decision-making process facing the firm?

First, consider the options facing the manager of the firm: saving domestically or save overseas. One could save $1 in an US treasury bond, and get an interest rate of iUS at the end of a year. Or one could take that one dollar, convert it into euros at the rate of (1/S) €/$, and get an interest rate of iEuro at the end of the year. Thus one has the following choices, with the following returns:

Save in US Save in the Euro Area ------------ --------------- (1+iUS) in $ (1+iEuro)/S in €

Presumably, if you are an American, you don’t really care how many euros you expect to have. Rather, you care how many dollars you expect to have, so your decision must also depend upon what you think the exchange rate will be at the end of the year (call this is Se+1). Hence, the relevant comparison is between the returns from these two choices:

(2), Pre announced horizontal band that is narrower than or equal to +/- 2% (3), De facto peg (4), Pre announced crawling peg (5), Pre announced crawling band that is narrower than or equal to +/- 2% (6), De facto crawling peg (7), De facto crawling band that is narrower than or equal to +/- 2% (8), Pre announced crawling band that is wide than or equal to +/- 2% (9), De facto crawling band that is narrower than or equal to +/- 5% (10), Moving band that is narrower than or equal to +/- 2% (11), Managed floating (12), Freely floating (13), Freely falling (14). 4 Frankel (2004). Other researchers have tabulated de facto exchange rate regimes, including Levy-Yeyati and Sturzennegger (2003), and Ghosh, Gulde, Ostry and Wolf (2002); the correlations are of a similar magnitudes.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Save in US Save in Euro Area ------------ --------------- (1+iUS) in $ (1+iEuro)Se+1/S in $

The comparison is a little difficult using these formulas. Using approximations, the comparison can be recast as follows:

Invest in US Invest in Euro Area ------------ --------------- iUS iEuro + (Se+1 - S)/S return on US return on Euro area bond plus bond change (depreciation) of $ against €

If all the saver cares about is the expected return, then if the left hand side is larger than the right hand side, then put the savings in the US. If the reverse is true, then put them in the Euro Area.

If everybody thought the same thing about what the exchange rate would be in a year, eventually all the financial capital would migrate to the same location, moving interest rates and the expected changes in the value of the currency until such point that the following equality held:

(10.1) ≡ ∆

This expression indicates that the interest differential equals expected depreciation. Consider the case where the US interest rate is 5%, the euro area interest rate is 3%; then expected depreciation should equal 2%.

This condition is called uncovered interest parity (UIP for short), and it is a type of “no arbitrage profits condition”. It's exactly because everybody doesn't think the same way that we have these vast amounts of trades in foreign currencies. Clearly some people think the dollar will lose more value over the year than the 2%, and they will move their money to the euro area. Those who think the dollar will lose less than 2% will be moving their money to the US. Uncovered interest parity is more a convenience than an empirical reality, as far as can be determined. Box 10.1 examines the evidence for and against UIP holding in the real world.

Since we don’t ever actually see the market’s expectations, the uncovered interest parity condition cannot be verified. However, market participants can enter into agreements to trade in the future. These contracts are called forward exchange contracts. In this case, the relevant

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

expression, in which case one can’t make higher certain returns saving in the US or in the euro area, is:

(10.2) /

Then the interest differential equals what is called the forward discount. This is another no arbitrage profits condition, called covered interest parity (CIP for short). Consider for example the case where the US interest rate is 10% and the Euro area interest rate is 7%. Then, in order for one to not be able to get a higher dollar return in one location versus another, the forward exchange rate for a trade one year hence must be 3% higher than the exchange rate today. Then, the one obtains 10% at home, or 10% by saving abroad (7% in interest, 3% in gain of the euro against the dollar).

One might decide to undertake this particular transaction which provides certain returns, instead of the other, because one doesn't like taking chances -- that is one is "risk averse".

When one undertakes this procedure, one certainly decreases the risk of bad outcomes. On the other hand, if the dollar loses its value by 5%, then one only obtains 10% return instead of the 12% return if one had taken the uncovered route.

These two conditions will only hold if governments do not put any barriers to the movement of your money, or financial capital. In the past, almost all governments put barriers to movements of capital, like restrictions on how much cash you can move in or out of the country. Currently, the extent of such barriers are very small for developed countries like the US, Germany, UK, and to a lesser degree, Japan.5 Hence, covered interest parity holds between such countries.6

For developing countries, however, like India, China, those in Africa, some Latin American countries, these restrictions are likely to be in place, and binding. Moreover, there is always the threat of the government imposing restrictions on the movement of savings in and out of the country (or even seizing it). In such cases, it would be highly unlikely that you would find covered interest parity conditions holding.

5 One will hear often discussions of the Eurodollar and Eurobond markets. These are offshore markets, not necessarily in Europe (the term arose because the first such markets were in Europe). The reason why these markets arose is because they could trade outside of the control of regulatory institutions in the various countries. For instance a British bank can take deposits in dollars, and pay in dollars. Because the bank is in London, it is not subject to the control of the American authorities. These markets originally arose in response to the aforementioned capital controls. They still exist because they circumvent all sorts of national banking regulations. 6 See Coffey, Hrung, Nguyen, and Sarkar (2009) and Griffoli and Ranaldo (2009) for a discussion of deviations from CIP during the financial crisis.

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

Box 10.3 Does Uncovered Interest Parity Hold?

It is hard to test uncovered interest parity since expected depreciation is central to the condition, but these expectations can’t be directly observed. All we observe are actual exchange rate depreciations. If expected and actual changes are similar, then we should expect to see the observations line up on a positive sloped line. However, if one plots actual exchange rate changes against interest differentials, as in Box Chart 10.2, one sees something much different.

Box Chart 10.2: Annualized 3 month depreciation of US dollar/euro exchange rate against US-euro area 3 month interest rates, in percentage points.

The slope is negative, rather than positive; this is not a finding specific to the dollar/euro exchange rate. It is also holds true for other currencies, and remains one of the most robust stylized facts in the literature.

There are several possible explanations for this finding. First, it might be that the participants in the foreign exchange market have systematically wrong expectations about future exchange rate changes.

-40

-20

-2 -1 0 1 2 3

interest differential

d e

p re

ci a ti

o n

US dollar/ euro rate, 1999-2010

slope = - 2.88

Chinn/Irwin International Economics, Chapter 10 (draft 9/11/2017) © Menzie Chinn

The first proposition is difficult to test, since we don’t directly observe expectations. However, several organizations survey individuals at major financial institutions involved in foreign exchange trading. Presumably these people’s views are representative of the relevant market participants. Statistical tests confirm that their expectations are biased – expectations of exchange rate changes are typically wrong, so much so that for the dollar-euro exchange rate, the dollar appreciates when it’s expected to depreciate.

Plotting expected depreciation against the interest differential leads to a substantially different picture (Box Chart 10.3). Then the observations line up much more in line with what’s expected.

-40

-20

-2 -1 0 1 2 3

forward premium (interest differential)

e xp

e ct

e d

d e

p re

ci a

ti o

slope = +0.97

US dollar/ euro rate, 1999-2010

Box Chart 10.3: Annualized 3 month expected depreciation of US dollar/euro exchange rate against US-euro area 3 month forward premium, in percentage points.

The regression slope of 0.97, which is statistically indistinguishable from one, is consistent with the view that investors do equalize expected returns – it’s just that

investors are not very good in making their expectations.78 This finding also holds for other exchange rates, such as the Swiss franc and Norwegian kronor.

An alternative, and not necessarily exclusive -- explanation for the finding that interest rates wrongly predict the direction of exchange rate changes is that some currencies are more risky than others. This could be the case In that case an exchange risk premium drives a wedge between exchange rate depreciation and interest rate differentials.

(10.3) ∗ ∆ ≡

The risk premium is the excess return on securities denominated in a given currency, taking into account expected changes in currency values. When uncovered interest parity holds, the risk premium is zero. But there is no reason that this need be true in the real world. It might be that each household and firm around the world holds a portfolio of assets that includes for instance US government bonds, Japanese government bonds, and German government bonds. If they think of these bonds as imperfect substitutes, so that for instance they have a preference for Japanese bonds, then when the expected dollar returns on German bonds exceeds that of Japanese bonds, everybody doesn’t dump the Japanese bonds in favor of the German bonds. They might try to increase their holdings of those German bonds, but still retain a few of the Japanese bonds.

In a world where government bonds are not perfect substitutes, the expected excess return on – say US bonds – might be there to compensate for the fact that the US government has issued more bonds than households and firms around the world desire (how much are desired could depend on how the currency and interest rates move together, or with consumption), as well as the extent to which those actors are risk averse. If agents are risk neutral, then the risk premium should be zero, and uncovered interest parity holds.

10.5 More on Exchange Rates

So far we’ve discussed exchange rates as the relative price of two currencies -- e.g. the US dollar and the euro -- and this is appropriate as this price (also known as a bilateral exchange rate) is the most commonly understood use of the term. However, sometimes we want to think of the exchange rate as the price of a

7 These data differ from the data used in Box Chart 10.1, which were dated at the end of each quarter. These data pertain to the third Thursday in the last month of each quarter. The variable that stands in the stead of the interest differential is the forward premium, which should equal the interest differential, but in practice differs slightly. Finally, the slope coefficient reported in Box Chart 10.2 is based on a regression which excludes the observation for December 2008, which pertains to the height of the global financial crisis. Including this observation results in a regression coefficient of 1.7, also statistically indistinguishable from the value of one. 8 These results are based on data drawn from Currency Forecasters Digest and FX4casts, and used in Chinn (2014). Earlier contributions in this vein were Dominguez (1986), Frankel and Froot (1987), and Froot and Frankel (1989).

number of several other currencies relative to say the US dollar. Or we want to think of the relative price of US goods relative to a foreign goods.

On the first point, it would be nice to have a summary measure of the value of a specific currency. However, constructing such a measure is complicated because bilateral exchange rates do not always move in the same direction. In figure 10.1 we can see that toward the end of 2012, the dollar remains stable against the pound, even as it appreciates against the yen. This is to be expected, as the exchange rate is the relative price of the two respective currencies, and each currency’s value will be driven by different factors.

How, then, to characterize for instance the movements of the currencies overall? One way is to calculate the value of a currency against the value of a basket of currencies. However, a simple average would not make sense. For instance, should the currency of Lithuania, which has very little economic interaction with the United States, have equal weight with the currency of Canada, a country that is the largest trading partner of the United States? Clearly, the answer is no. A typical practice is to weight the currencies in the basket by the amount of trade conducted between the home and foreign countries. For instance, if the US conducted 1/3 of its trade with the Euro area, 1/3 with China, and 1/3 with the rest of the world, the weights for the euro would be 1/3, for the yuan 1/3, and for the sum of the other currencies, 1/3.9

Chart 10.4 presents the values of the currencies examined in Figure 10.1. Note that in this case, the rise in each series is interpreted as an increase in the value of the indicated currency (i.e., Deutsche mark, euro and Japanese yen).

9 The construction of the indices is slightly more complicated than laid out here. Instead of an arithmetic weighted average, the indices constructed by statistical agencies use a geometric weighted average. Chinn (2006) describes the construction of the various types of effective exchange rate indices.

Chart 10.4: Trade weighted nominal currency values for Deutsche mark (blue), British pound (green) and euro (red) against a basket of currencies, weighted by trade flows, 2005=100. Higher values indicate a stronger currency.

So far we have been discussing what are called nominal exchange rates, as they pertain to the tradeoff between paper assets (currencies). However, for economic transactions, we are often concerned with real magnitudes; is there a real counterpart to these nominal exchange rates. In fact there is, conceptually. This is most easily understood by explicitly considering the units. Let’s consider for instance the nominal exchange rate between the US and the euro area, from the perspective of a US resident.

Which is the number of dollars required to obtain a single euro. The real exchange rate measures the rate of exchange of real goods. Define the prices of widgets in the US and euro area as:

The real exchange rate is thus given by:

(10.3)

≡ ∗

100

120

140

160

180

1970 1975 1980 1985 1990 1995 2000 2005 2010

DEM EUR

GBP

Currency value against basket of currencies

In other words, the real exchange rate is the number of US widgets one has to give up to obtain one single foreign widget.

If US widgets traded off one for one with Euro area widgets, then q=1, and the price of a US widget in USD should equal the price of a Euro area widget in USD terms.

(10.4) ∗

In a world where US and Euro area widgets were exactly the same, and there were no impediments to trading those widgets across borders and prices could be easily adjusted, a reasonable value for q would be 1. However, the real world there are barriers to trade and firms do not seem to adjust prices moment by moment; as a consequence, q might not always equal 1, although it would seem reasonable that over time q would gravitate toward that value.

In general, we don’t care about widgets, but rather bundles of goods and services. One measure of prices of bundles of goods purchased by consumers, namely consumer price indices (CPIs). Figure 10.3 shows the bilateral real exchange rates corresponding to the nominal rates in Figure 10.1, deflated using CPIs.

0.0

0.4

0.8

1.2

1.6

2.0

2.4

70 75 80 85 90 95 00 05 10 15

EXP(QGR) EXP(QEU) EXP(QUK)

USD/DEM USD/EUR

USD/GBP

Real exchange rates against US dollar

Chart 10.5: Real exchange rates for Deutsche mark (blue), British pound (green) and euro (red) against USD. Exchange rates deflated using consumer price indices (2010=100). Higher values denote a weaker dollar against the foreign currency (a depreciation).

Each of these series can be interpreted as the number of bundles of US consumer goods and services required to purchase a single bundle of foreign consumer goods and services.

One interesting aspect of these series is that they exhibit less pronounced trends than the corresponding series displayed in Chart 10.1. This contrast makes sense, given our discussion of q. Chart 10.1 pertains to the relative price of pieces of paper, namely currencies. Those can drift far apart as demand and supply of pieces of paper vary. Chart 10.5 depicts the relative price of bundles of goods and services. If an American bundle of goods and services is similar to a bundle of European goods and services, then the

Finally, it is of interest to consider how the bilateral real exchange rates against USD compare against the real trade weighted USD exchange rate. Chart 10.6 illustrates that the trade weighted real exchange rate (defined up as a depreciation) is a useful way to summarize overall movements in a currency.

Chart 10.6: Log exchange rate for euro (dark blue), Japanese yen (red) and Chinese yuan (light green) against USD, and log real trade weighted exchange rate for USD (bold gray), normalized to 2002M01=0. Exchange rates deflated using consumer price indices. Higher values denote a weaker dollar (a depreciation).

When the USD/EUR real exchange rate depreciates, the trade weighted real exchange rate tends to depreciate, since the US has a large amount of trade with the euro area. However, the two series don’t move in lockstep, exactly because the Japanese yen (as well as the Chinese yuan) is moving in different directions at various points in time.

-.1

1996 1998 2000 2002 2004 2006 2008 2010 2012

USD/EUR

USD/JPY

Trade w eighted USD

Real exchange rates, in logs 2002M01=0

USD/CNY

Summary Points

1. The nominal exchange rate is the price of foreign currency, in terms of home currency.

2. Under a floating exchange rate regime, supply and demand determine the price. Under a fixed exchange rate regime, the central bank must freely buy and sell currency in order set the price at the official exchange rate.

3. Under a floating (fixed) rate regime, a rise in the price is called a depreciation (devaluation), while a fall is an appreciation (revaluation).

4. Most exchange rate regimes in existence incorporate intermediate degrees of flexibility.

5. The forward exchange rate, the current (or spot) exchange rate, and interest rates are linked together.

6. The real exchange rate is the relative price of a bundle of foreign goods, in terms of bundles of home goods.

7. The trade weighted value of a currency is the value of the currency relative to a bundle of foreign currencies, where the weights are based on trade flows.

References

BIS, 2013, Triennial Central Bank Survey: Report on global foreign exchange market activity in 2013: preliminary global results (Sept. 2013).

BIS, 2010, Triennial Central Bank Survey: Report on global foreign exchange market activity in 2010 (Nov. 2010).

Chinn, Menzie, 2006, “A Primer on Real Effective Exchange Rates: Determinants, Overvaluation, Trade Flows and Competitive Devaluations,” Open Economies Review 17(1) (January): 115-143.

Chinn, Menzie, 2014, “(Almost) A Quarter Century of Currency Expectations Data: Interest Rate Parity and the Risk Premium,” mimeo (September).

Chinn, Menzie, 2015, “Emerging Market Economies and the Next Reserve Currencies,” Open Economies Review 26(1): 155-174.

Chinn, Menzie and Jeffrey Frankel, 2007, “Will the Euro Eventually Surpass the Dollar as Leading International Reserve Currency?” in R. Clarida (ed.), G7 Current Account Imbalances: Sustainability and Adjustment (U. Chicago Press), pp. 285-322.

Coffey, Niall, Warren B. Hrung, and Asani Sarkar, 2009, "Capital constraints, counterparty risk, and deviations from covered interest rate parity." Federal Reserve Bank of New York Staff Reports 393.

Dominguez, Kathryn. 1986."Are Foreign Exchange Forecasts Rational? New Evidence from Survey Data?" Economics Letters 21: 277-82.

Frankel, Jeffrey, 2004, “Experience of and Lessons from Exchange Rate Regimes in Emerging Economies,” Monetary and Financial Integration in East Asia: The Way Ahead, edited by Asian Development Bank (Palgrave Macmillan Press, New York), vol. 2, 91-138.

Frankel, Jeffrey and Kenneth Froot. 1987."Using Survey Data to Test Standard Propositions Regarding Exchange Rate Expectations." American Economic Review 77(1) (March): 133-153.

Froot, Kenneth and Jeffrey Frankel. 1989."Forward Discount Bias: Is It an Exchange Risk Premium?" Quarterly Journal of Economics. 104(1) (February): 139-161.

Ghosh, Atish Rex, Gulde, Anne Marie, and Wolf, Holger C., 2002, Exchange rate regimes: choices and consequences (Vol. 1). The MIT Press.

Griffoli, Tommaso M., and Angelo Ranaldo, 2009, Deviations from covered interest rate parity during the crisis: a story of funding liquidity constraints. Working Paper, Swiss National Bank.

Levy-Yeyati, E. and F. Sturzenegger, 2003, “To Float or to Fix: Evidence on the Impact of Exchange Rate Regimes on Growth,” American Economic Review 93 (4) (September).

Reinhart, Carmen M., and Kenneth S. Rogoff, 2004, "The Modern History of Exchange Rate Arrangements: A Reinterpretation," The Quarterly Journal of Economics 119 (1): 1-48.