Saturday, September 05, 2009

Implications of a Doomed Dollar

My friend Jeff Hummel has been arguing for some time that the U.S. is eventually going to default on its debt. Central to his argument is the fact that an increase in the perceived risk of default increases the risk premium the U.S. has to pay to lenders. Since the debt is large (and rapidly getting larger) an increase in the interest rate the U.S. has to pay results in significantly increased budgetary problems, which results in an increased risk of default, which results in a further increase in the risk premium—a positive feedback mechanism. I do not know if his prediction is correct, nor do I know whether, if it is, default would be via repudiation or inflation, but it is at least an interesting argument.

I was reminded of it recently when I received an Email from The Motley Fools, investment advisers who for some reason have me on their mailing list. The title is "Read This Because the Dollar is Doomed." The authors argue that the U.S. trade deficit, the U.S. budget deficit, and the willingness of the U.S. to pay for things by printing money all threaten the dollar. They conclude that "This should be worrisome news if you earn a dollar-based salary, keep a dollar-based bank account, or invest in dollar-denominated U.S. stocks and bonds. Why? Because as the dollar declines in value, so too will all of your earnings, savings, and investments. And that's scary stuff." They argue that the solution is to invest in "stocks that do business in other currencies ... and specifically in currencies that you suspect will rise against the dollar over time."

It is possible that the dollar is doomed, but their explanation and advice confuse two different issues—price levels and exchange rates. Inflation due to too much money creation is a problem if you have assets whose value is fixed in dollars, such as T-bills. But it isn't a problem for assets whose value is merely measured in dollars, such as U.S. stocks. If all prices double, the price of Apple computers doubles too, as does the value of Apple stock. Some companies will do better in an inflation than others, in part because some companies have assets or liabilities whose dollar value is fixed. But that has nothing to do with whether the company's stock is dollar denominated.

Nor does it have anything to do with whether the company does business in other currencies. If the value of the dollar drops and the value of the Euro doesn't, then the dollar value of a company that does business in Euros will go up. But so will the dollar value of a company that does business in dollars. The Euros the company is earning exchange for more dollars than they used to, but those dollars are worth less.

There is, however, a different sense in which the dollar might be doomed, with different implications for investors.

Suppose the U.S. price level stays the same but the exchange rate between dollars and other currencies falls. A dollar will still buy the same goods and services in the U.S. as before, but not as many Euros or Yen or Rupees.

Why might that happen? The market exchange rate between the Euro and the dollar is the price at which the number of dollars that people want to sell for Euros equals the number that other people want to buy. One reason to trade Euros for dollars is in order to buy goods in the U.S. and ship them back to Europe, and similarly in the other direction. If that is all that is happening the exchange rate ought to reflect the purchasing power of the currencies, with some complications due to the fact that not all goods and services play a role in international trade.

Another reason a European might want dollars, however, is in order to buy T-bills—more generally, in order to buy U.S. capital assets. If the U.S. is running a large budget deficit, as it is, and if much of that money is being borrowed from foreigners, as it is, then a lot of dollars are being bought in order to lend them to the U.S. government. That additional demand bids up the price of the dollar in exchange markets. The capital inflow appears in the statistics as a trade deficit, since some of the foreign goods the U.S. is importing are being exchanged for capital assets which remain in the U.S. rather than for export goods that don't.

Suppose foreigners decide that lending money to the U.S. government is no longer prudent. That part of the demand for dollars vanishes. The price of the dollar measured in Euros or Rupees falls. Some of the foreigners who have lent money to the U.S. decide it was a mistake, cash in their T-Bills for dollars, and trade those dollars for Euros or Rupees. That increases the supply of dollars on the foreign exchange market, driving the dollar down even farther. The Motley Fool isn't entirely clear about which sort of decline of the dollar he is talking about but one could, with a little effort, take his references to the trade deficit and the budget deficit as suggesting some mechanism along these lines.

What are the implications if the dollar maintains its value domestically but falls in its exchange rate? Americans who hold assets with fixed dollar values are unaffected, except to the extent that they want foreign currency, perhaps for a Paris vacation. Americans who hold stock in foreign companies, or in U.S. companies whose income is largely in foreign currencies, benefit, since their stock is now worth more dollars. The losers this time are foreigners who made the mistake of holding U.S. assets.

It is, perhaps, unreasonable to expect investment advisers to not only understand economics but to explain it correctly to their customers. On the other hand, since which part of their advice one ought to follow depends on what sort of doom the dollar is facing, it would be nice if they at least tried.


Anonymous said...

Is it really possible, in a world as connected as ours, that "U.S. price level stays the same but the exchange rate between dollars and other currencies falls"? It's hard to find goods in the U.S. market that aren't made (at least in part) abroad. So a drop in the value of the dollar implies a rise in domestic prices.

Skip said...

Sure it is. Around 2000 the dollar fluctuated around parity with the euro. It bottomed (so far) at almost $1.60 to the euro. Did we see anywhere near that much swing in US prices? As best I can tell prices in the US on most goods didn't change other than normal inflation.

On the other hand, I have a small collection of ancient coins I bought, mostly around then, and the prices on those have pretty much fluctuated in line with the dollar/euro exchange rate. So some things will definitely increase in price.

Anonymous said...

Since the government's debt is in its own fiat currency, the fate of the debt and currency are closely linked. Although depressions decrease tax revenue, they are good for money, and hence good for government debt. Thus a depression isn't a trigger for default.

Anonymous said...

Anonymous, I'm not sure I understand. A depression tends to lead to deflation, i.e. dollars getting more valuable relative to domestic goods and services, which means that the government suddenly owes more goods and services than it did before.

Deflation is good for people who HAVE money, and for people who ARE OWED money, but very bad for people (or governments) who OWE money.

Eric Hill said...

I have trouble with the unqualified assertion that stock prices follow (upward) inflation. There's at least one source,, that shows them to be frequently anticorrelated. Otherwise I appreciate the discussion of inflation vs. exchange rates.
(I think repudiation would make it much more difficult to do normal business in the future than inflation, even though that's not at all satisfactory either.)

Anonymous said...

"Deflation is good for people who HAVE money, and for people who ARE OWED money, but very bad for people (or governments) who OWE money."

Deflation is good for people who have money, true. It is not good for people who are owed money, because of credit risk. The one exception is government bonds, because if the government can't get the money from taxes, it can print it. Deflation is not a problem for governments which owe their own fiat money.

Anyway, having gone and read Jeff Hummel's piece, I see that I'm arguing against a claim he didn't make. His argument isn't that an economic problem will trigger a default. Rather, he predicts that the government will choose default rather than a tax increase or benefit cuts to deal with the long term funding problems of Medicare and such. I doubt it, but it's possible.

123 said...

Unexpected large inflation acts as a redistributive tax, reduces the long term growth of real output, increases risk aversion. The real real price of stocks is reduced via increased risk premiums and lower income. So the argument that stocks with domestic customers should be avoided have some merit.

dWj said...

A drop in the value of the US dollar against other currencies will tend to trigger inflation, though not one-to-one. I would expect that what they had in mind was indeed your treasury-selling scenario, with ensuing inflation. It makes sense that stocks would be denominated in real terms, so that they would rise with inflation, and my understanding is that the empirics are consistent with this idea with the caveat that one needs to control for a risk premium that is likely to rise when inflation does so; this may make stocks a bad hedge in the short run but a good hedge in the long run.

I think the problem, though, is that depreciation-induced inflation will be concentrated in tradable goods. Stock in a US company selling exclusively to Americans won't help hedge inflation from depreciation; stock in a company selling to foreigners will.

(I saw the Motley Fool article before I saw your post, and when I saw "dollar-denominated stocks" I thought, "I'm not even quite sure what that means". I suppose I buy produce in dollars, but I don't think of them as "dollar-denominated green peppers".)

montestruc said...

I have been concerned about this for some time. However the firm I work for does much of it's business exporting engineering expertise, and custom designed equipment to corporations and governments in other nations. Weakness of the dollar tends to make us more competitive in price.

My concern however is that we quote in US$, and our contracts last typically 2-4 years. One can see how if US inflation took off that could put us in a bind.

Anonymous said...

The most likely doom for the dollar is all of the above, and none of the above, therefore specifying it in detail appropriate for normal times is pointless. When the dollar collapses, things will be abnormal.

The analysis you ask for is appropriate for small changes in normal times, but the doom that I think most likely is people ceasing to use the federal reserve dollar at all, with crisis setting in around 2020 or so, and actual dedollarization setting in around 2025 or 2030

Anonymous said...
This comment has been removed by a blog administrator.
Mercy Vetsel said...

Actually the U.S. will never technically default as long as the debt is denominated in dollars. That's the reason the rate for treasuries is called the risk-free rate.

It's not because it has no risk, but because it has no default risk.

Inflation? Well it seems pretty obvious to me that the government will inflate the debt away. That's what they did after WWII.


albatross said...

Won't the big impact on most Americans be the price of imports?

John T. Kennedy said...

"Inflation due to too much money creation is a problem if you have assets whose value is fixed in dollars, such as T-bills. But it isn't a problem for assets whose value is merely measured in dollars, such as U.S. stocks."

Runaway inflation in the U.S. couldn't negatively impact U.S. stocks relative to foreign stocks?

That seems implausible.

John T. Kennedy said...

I agree that speaking of "dollar denominated" assets makes little sense; all assets have a denomination in dollars.

Yet I think U.S. companies will do disproportionately badly if the dollar fails, which means you'll be better off invested in foreign assets, which I think is the gist of their advice.

VangelV said...

It is, perhaps, unreasonable to expect investment advisers to not only understand economics but to explain it correctly to their customers.

Why not? How can one really be an investment advisor unless s/he understands economics well enough to handle the macroeconomic risks? I do not think that the lack of economic education is the problem. The actual problem is the type of economic education provided by most of the Western universities. Sadly, most students learn Keynesian and monetarist theory and ignore the Austrian school entirely. As such, they are ill prepared to understand how the economy really works and are prone for the very big errors that usually wind up doing a lot of damage to portfolios. In my own case I avoided economics in university because I could never tolerate lecturers that were obviously over their heads but too arrogant to figure it out. The University of Toronto, where I studied engineering, was primarily teaching failed Keynesian ideas, which was of no interest to me whatsoever.

Sadly, I first really learned about Mises and Hayek when I was working in China. Their ideas were used by local economists to defend economic liberty in a debate with my American and Canadian colleagues, who were arguing for more central planning and meddling by governments.