Thursday, October 14, 2010

Currency Wars

There has been a lot of fuss recently about exchange rates - the relative prices at which currencies trade on foreign exchange markets. I find the whole issue very puzzling, and this is my attempt to sort this out.

Let's start at the local level. In the United States (and probably in other countries too), there is a local currency movement. The most prominent (and on some level, successful) example is Ithaca hours. In Ithaca, New York, a group of what are apparently local merchants issues pieces of paper, with the members of the group agreeing to accept this currency in exchange for goods and services. The unit of currency is an "hour," and the exchange rate of hours is fixed (apparently by decree) at 1 Hour per 10 U.S. dollars. Apparently these notes are not redeemable in anything, in contrast to private currency systems that existed historically, for example before the Civil War in the United States. It is not clear who gets the seigniorage (the initial benefit from the issue of the currency), but I'm guessing that, if one signs up as a member, one gets in return some quantity of hours to spend.

What motivated these Ithacans to start up their currency system? This (I guess we would call it a graphic short story) tells us. Note in particular points 1 through 7 starting in the middle of the page. It should be clear that the hours-issuing Ithacans did not attend classes in conventional modern economics. The theory underlying their currency system is in part related to social credit and Marx's labor theory of value, with some wishful thinking thrown in for good measure. The wishful thinking relates to the ideas that exchange using hours can somehow enforce a minimum wage of $10.00 per real hour (i.e. there is Ithaca hour illusion), and eliminate cutthroat capitalism, thus making the economy somehow more friendly. However, for later use, note three key ideas in their story: (i) money is not neutral: more units of it in circulation increases local employment; (ii) there is a protectionist element: form a local club, which promotes trade among members of the group, the corollary being that there is less trade with the rest of the world.

Now, going beyond the local level, historically governments were motivated to establish national currencies in part to generate revenue. If it is very costly or impossible to collect taxes to finance government spending, it can be efficient for a government to establish a monopoly over the issue of money. The government can then exploit this monopoly power, particularly in wartime, by printing money to finance its expenditures. However, the establishment of a national currency could be motivated by some of the same ideas that got the hours-issuing Ithacans excited. If a government can successfully establish its currency as the most widely-accepted medium of exchange within its borders, this serves a protectionist role, in that it makes exchange between domestic residents and the outside world difficult. To export or import goods requires that I convert domestic currency into whatever medium of exchange is used outside my own country, and the government can make this even more difficult by throwing sand in the gears of the foreign exchange market. It is this idea that motivated the formation of the European Monetary Union, under which the medium of exchange in most of Europe became the Euro, thus lessening frictions associated with trade in goods among European countries.

Now, let's move to the current state of the world. Some people, including Tim Geithner, IMF officials, look at the world this way:

1. There are global "imbalances," i.e. some countries are running current account surpluses (e.g. China and Germany), and others are running current account deficits (e.g. the United States).

2. The way to cure these imbalances is to have the exchange rates of the surplus countries rise, and those of the deficit countries fall.

3. To the extent that surplus countries engage in policies that prevent exchange rate appreciation, this is bad for the world, as it perpetuates the imbalances.

What happens when a country runs a current account surplus? Roughly, it is selling more goods and services to the rest of the world (exports) than it is buying from the rest of the world (imports). The rest of the world has to be paying for this net outflow of goods and services with an inflow of assets to the surplus country. A country with a current account surplus is effectively accumulating claims on the rest of the world - it is a net lender in the world economy. Similarly, a country with a current account deficit is a net borrower in the world economy.

Now, there are plenty of good reasons for a country to be running a current account deficit and borrowing from the rest of the world. First, a country may have a period when its productive capacity is temporarily low. For example, there could be a widespread crop failure, and it can then make sense to borrow temporarily from the rest of the world to finance imports of food. This is the consumption-smoothing motive for a current account deficit. Second, a country may have high-return domestic investment projects, and there may be insufficient domestic savings to finance that investment. A country can borrow on world markets to finance the investment projects, and effectively repay the loans in the future when these projects pay off (when the country will be running a current account surplus). This is the borrowing-for-investment motive. Third, a country may choose to finance government spending by issuing government debt to foreigners. Indeed, this may be an efficient method of financing, relative to the alternatives. This is the government-borrowing motive.

Since there are at least three good reasons for running a current account deficit, there are also at least three good reasons for running a surplus, and we can imagine an efficient state of affairs where the countries of the world are never in "balance," where balance is a state where the current account surplus is zero in all countries. Just as with particular households, where there is an ebb and flow to borrowing and lending and the quantity of debt over the household's lifetime that allows it to deal with fluctuations in income and investment opportunities, so it is with countries. What are Treasury Secretary Tim and the IMF all wrought-up about then? Possibly what they have in mind is that current account surpluses and deficits appear to be quite persistent. For example, the United States has run a current account deficit for most of the period since the early 1980s. Possibly the holders of the debt of deficit countries will ultimately come to doubt that these countries are capable of paying off their current stock of debt (as has already happened to some countries), and creditors will demand high premia to lend to these countries or well cease lending entirely.

If this is the case, the "imbalance" problem solves itself, through the actions of international lenders. But maybe this solution is too costly. Sovereign debt problems (as in the recent case with Greece) develop suddenly, and entail potentially severe short-term disruptions. Another potentially less costly approach for a country that is having trouble servicing its debt is to implicitly default. This can be accomplished through monetary expansion. An easy monetary policy increases domestic prices, thus reducing the real value of the government's debt in terms of domestic goods, and the exchange rate tends to depreciate. No matter what foreign governments do, the deficit country's debt is worth less in terms of foreign goods as well. Of course the long-term cost of implicit default is that international lenders will now trust you less - keep inflating away your debt, and you pay a high premium on world credit markets, so there is really no easy way out.

Now, the way the current "imbalance problem" is framed, by Geithner for example, is as if the problem of the US as a debtor nation was created by the thrifty habits of the creditor Chinese. Further, the problem is almost always described in terms of the relative prices of goods on international markets and the effect of those prices on the current account balance. This seems odd, particularly as the pattern of current account balances across countries is best explained in terms of how credit markets work domestically and internationally. From Geithner's point of view, a current account surplus implies that the relative prices of goods are somehow wrong. China is selling us too much stuff and buying too little of our stuff, according to Geithner, and the solution to this problem is to increase the relative price of the stuff that China sells relative to the price of stuff that China buys. What better way to do that than to have China allow its currency to appreciate?

This makes no sense. Suppose, for example, that we had balance of payments accounts by state for the United States. Suppose that those accounts indicated that California was running a persistent current account surplus relative to New York. Would we think it a serious problem that exchange rates are fixed within the United States, and argue that a solution to our "imbalance problem" is to have individual state currencies, so that we could allow the California currency to appreciate relative to New York currency?

The ideas motivating Geithner seem as misguided as, and indeed are similar to, those motivating the issuers of Ithaca hours. The surprising thing is that the issuers of Ithaca hours have never contemplated a devaluation, which according to Geithner logic would steal jobs from Syracuse, Elmira, or possibly Binghamton. Of course this would lower the minimum wage in Ithaca, according to this, which may explain why this has not been attempted.


  1. Why does the exchange rate matter?

    We should be care about the relative prices of goods. The prices of Chinese goods will simply reflect the value of the currency. Am I wrong; does the exchange rate affects anything but the units of measurement?

  2. No, the problem, as described by virtually everyone except you, is that the Chinese government is accumulating large amounts of dollar-denominated assets, which prevents the Chinese currency from rising relative to the dollar and thus creates an artificial price advantage for Chinese goods. I would suggest addressing the argument as it is made, rather than attacking Ithaca, entertaining as that case may be.

  3. Professor,

    Thanks for the tip on the Math preparation, it was very helpful. Could you expand upon the notion that China is "manipulating" its currency through open market operations. Thanks.

  4. "No, the problem, as described by virtually everyone except you, is that the Chinese government is accumulating large amounts of dollar-denominated assets, which prevents the Chinese currency from rising relative to the dollar and thus creates an artificial price advantage for Chinese goods."

    Yes, this is what a country does when it fixes its exchange rate with respect to another currency. There is nothing "artificial" about it. Again, it's like saying a state in the US has an artificial price advantage over other states if it is running a current account surplus relative to the other states. Nonsense!

  5. "We should be care about the relative prices of goods. The prices of Chinese goods will simply reflect the value of the currency. Am I wrong; does the exchange rate affects anything but the units of measurement?"

    Money neutrality implies that the exchange rate should not matter ultimately for relative prices. However, there is considerable empirical evidence to suggest that the variability in real exchange rates over the short run is driven primarily by nominal exchange rate variability. Some people think that this is because of Keynesian sticky prices, and there are other potential explanations like market segmentation. If it's sticky prices, that would suggest that a Chinese currency appreciation would knock relative prices out of whack, which would be bad. Geithner is saying that the Chinese current account surplus tells you the prices are out of whack and need to be adjused, which I think is wacky.

  6. I'm not quite sure what it means for nominal exchange rate variability to drive short run real exchange rate variability. The natural question then is, what is driving nominal interest rate fluctuations?

    The reason I ask about relative prices is I hear arguments that Chinese exchange rate policy has some adverse effect on wage rates in China, which also seems like nonsense.