April 14, 2008
The G-7 Communique and the Dollar
Figure 1: Log nominal value of US dollar against other major currencies. NBER defined recession dates shaded gray. Source: Federal Reserve Board via FRED II.
Lagarde Says G-7's Dollar Shift on Par With 1985 Pact (Update2)
By Christopher Swann and Kathleen Hays
April 14 (Bloomberg) -- French Finance Minister Christine Lagarde said investors haven't grasped the magnitude of the Group of Seven's shift in stance on exchange rates, likening its significance to the 1985 Plaza Accord.
"It's a strong statement which I am not sure the markets have yet fully understood and appreciated," Lagarde said in an interview on Bloomberg Television in New York.
Lagarde's remarks indicate she may be frustrated that the dollar failed to rally today after the G-7 warned that recent "sharp fluctuations" in exchange rates risk hurting the global economy. The group of seven major industrial nations issued their statement after the dollar slid 8 percent against the euro and 6 percent versus the yen since they last met in February.
The April 11 statement was "not very different" from the importance of the 1985 Plaza Accord, she said. The Group of Five at the time agreed to "coordinated intervention" to drive down the dollar. The G-5 then included the U.S., U.K., France, Germany and Japan. It later expanded to add Canada and Italy.
The article continues, documenting widespread skepticism among financial market participants. This view is summarized by this quote from today's IDEAGlobal newsletter (April 15):
The FX markets digested the latest G7 statement with relative ease on Monday, with EUR/USD eventually retracing all its losses posted soon after the statement was first issued. The wording changed significantly, but there no little [sic] meaning behind the change of syntax. The more things change, the more they stay the same. The environment still remains one of very negative sentiment towards the greendback...Monday's US economic outlook release calender did nothing to change this outlook....This supports the notion US continues to be in a recession, leaving the US vulnerable to further losses.
I think this quote does provide several reasons for skepticism, but also highlights why these pronouncements could matter.
First, reasons for skepticism. Policymakers, especially finance ministers, have limited means to affect exchange rates directly. In the case of the United States, the Treasury can intervene in foreign exchange markets, purchasing foreign currency with dollars, and sterilizing the increase in the dollar money supply by selling Treasuries.
Admittedly, the evidence that sterilized intervention works is actually more mixed than is commonly allowed. For certain, there do not seem to be persistent effects on the level of exchange rates from sterilized intervention for major currencies. This is true even when the intervention is in the amount of hundreds of billions of dollar, as in the Japanese experience during 2003-04 ($315 billion to be exact). But they do seem to exist. , , .
When effects of intervention have been identified, they typically (although not always) seem to be of a short term nature, affecting the direction of the exchange rate's movement over a few days.
On the other hand, China's ability to manage the exchange rate by accumulating hundreds of billions of dollars of reserves has led to some revision. But of course, this is a case where the intervention is not fully sterilized, and is supported by a panoply of capital controls. Whether such an effect would apply for developed country currencies depends upon (in part) whether a portfolio balance channel exists (i.e., whether bonds denominated in different currencies are perfectly substitutable), even if the intervention is coordinated., .
Hence, I think this paragraph from Owen Humpage's "Government Intervention in the Foreign Exchange Market" (2003) summarizes the profession's thinking on the subject, insofar as intervention on developed country currencies is concerned:
Sterilized intervention affords monetary policy makers a means of occasionally pushing an exchange rate in a desired direction. The alternative level then serves as a new starting point for a random walk process compatible with existing fundamentals. The empirical support for this conclusions seems consistent with the idea that monetary authorities periodically possess better information than other market participants and, in these instances, can sometimes can affect market expectations through intervention. This description does not preclude intervention as a signal of future monetary policy, but interpreting intervention as solely or even primarily such a signal is probably wrong. The likelihood that a given intervention will have the desired effect increases if the transaction is large and coordinated with the foreign monetary authority whose currency is involved.
Nevertheless, because sterilized intervention does not affect market fundaments, it does not afford monetary authorities a means of routinely guiding their exchange rates along a path that they determine independent of their monetary policies. While monetary authorities in large developed countries certainly can affect nominal exchange rates through non-sterilized foreign exchange intervention, doing so either will conflict with their domestic policy objectives or it will be entirely redundant to open market operation in domestic securities. The outcome depends on the nature of the underlying economic shock to their exchange market.
This highlights the one point where I think G7 pronouncements can matter. If policy makers view the exchange rate as a variable of primary interest, then they will try to alter the fundamentals (tax revenues and government spending, when the finance ministers make the pronouncements). Even if these fundamentals don't change at the time of announcements, if the commitment to change the expected path in the fundamentals is credible, then the exchange rate should change at the time of the announcement.
However, no such commitment seemed to have been made, at least on the fiscal side. What about the monetary side? To the extent that the central bank governors (including Bernanke) were also there, then there could be some informational content in the communique.
Yet, this would require more "weight" on the exchange rate (separate from its influence on output) than has previously been evident in the Fed and ECB reaction functions (estimated weights are fairly small, e.g., , and it is hard to determine whether that is the direct effect, or the exchange rate gap is standing in for an effect working through aggregate demand; see the framework in Chinn and Meredith (2004)). That's possible. Likely? Well, these are extraordinary times.
Bottom Line: If the Fed perceives that the dollar has dropped far enough to keep the U.S. economy out of a deep recession, and the ECB perceives the euro has risen enough to adversely impact euro area economic activity at a time of a pronounced slowdown (see here), then "the stars will be aligned" for a change in the path of (monetary) fundamentals, and hence the value of the dollar. Until then, the dollar will be buffeted by news about output, incomes, and credit markets (a different portfolio balance channel ).
Posted by Menzie Chinn at April 14, 2008 09:30 PMdigg this | reddit
Listed below are links to weblogs that reference The G-7 Communique and the Dollar:
» G7-April-2008-Meeting and Plaza-Accord from Kasinomics
French Finance Minister Christine Lagard has compared the recent FSF-report released by the G7 Finance Ministers last week to the Plaza Accord (as quoted by Christopher Swann in an article at Bloomington): The April 11 statement was “not very dif... [Read More]
Tracked on April 15, 2008 08:09 AM
So sterilized interventions have no permanent effect on exchange rates, while non-sterilized do? Is this correct?
Posted by: David at April 15, 2008 06:01 AM
If the Euro becomes the reserve currency would that mean the US would be paying its debts in Euro rather than dollars?
Posted by: JS at April 15, 2008 06:01 AM
David: All asset based models of exchange rates I am aware of imply changes of money stocks -- which would attend non-sterilized intervention -- imply changes in the nominal exchange rate, holding all else constant. Some models can imply permanent changes in nominal rates due to even sterilized intervention, but the empirical evidence is mixed.
JS: Not necessarily. The US dollar could lose out on being the key reserve currency, but the US Government might still be able to borrow in domestic currency (this is related to the "original sin" literature pertaining to emerging market borrowing). However, the terms at which the US Government -- and private entities -- borrow might become less advantageous (see "exorbitant privilege" discussions in this blog).
Posted by: Menzie Chinn at April 15, 2008 06:27 AM
I think it's pretty obvious that what will change the fortunes of the dollar is, on the American side, an end to war and the deficits that go with it. That and ending our promotion of deindustrialization through tax and trade policy.
Europe's real problem is with the Asian currencies, although I'm sure they'd like a 20% adjustment against the dollar.
Posted by: Charles at April 15, 2008 07:36 AM
Nice graph, Menzie. It really is an education for me. I had no idea the heights that the dollar reached in '84. It also certainly puts the current dollar value in perspective. I do wonder how the Euro is accounted for, since it did not exist for most of the time series of that graph.
I think it's pretty obvious that what will change the fortunes of the dollar is, on the American side, an end to war and the deficits that go with it.
That's not how I read that graph. The dollar side started in '02.
Posted by: Buzzcut at April 15, 2008 08:03 AM
Fundamentally, I believe that the greenback has a considerable distance to fall until it stabilizes. The current account deficit is roaring along at $739 billion/year (4.7% / GDP), and the federal budget has a 2.4 % / GDP deficit. This red ink has persistent for about 30 years. Private investors have little reason to hold long term dollar denominated assets under these conditions, unless U.S. interest rates rise substantially. 3 month T-bills at 2.08% and 10 year T-bonds at 3.46% will not cut it when oil has increased 96%, food by 61%, and industrial commodities by 26% against the dollar in the last year !
From a policy standpoint, the low FED interest rates will surely result in an accelerated exchange rate loss for the dollar. As I recall, trade deficits adjust as -7% exchange rate per 1% of (current account deficit) CAD/GDP improvement, or 4% reduction in GDP per 1% CAD/GDP improvement. Exactly how this will work out is debatable, but I would expect, in the best of circumstances, that the US GDP will decline about 4% and the dollar exchange rate to decline about 26% during the current financial correction, in order for the Current Account to come into balance.
Jawboning by the ECB or the FED about defending the dollar are more for political consumption than realistic policy. It is inevitable that the dollar devalues further.
Posted by: MarkS at April 15, 2008 08:57 AM
Buzzcut: Thanks, glad you found the graph illuminating. Note that this is a time series plot of the nominal value of the dollar. For the real value, see this post, and for an alternative measure (WARP), see this post.
Posted by: Menzie Chinn at April 15, 2008 09:16 AM
Since almost all of our trade deficit comes from two sources : Oil Imports and China, won't we solve the trade deficit problem if we just did two things :
1) A tax on imported oil, to reduce US consumption while forcing importers to shoulder some of the burden.
2) Keep pressuring China to revise the RNB upwards until it is at fair value.
Won't these two measures be enough, if they were implemented?
Posted by: GK at April 15, 2008 12:50 PM
Note that this is a time series plot of the nominal value of the dollar.
Okay, so not quite as bad as it looks. We're on the verge of being in uncharted territory, but not quite there yet.
That's my impression, especially as the yen came back over 100 and has stabilized at 101. We'd still need a significant strengthening to get anywhere near the record of 80 to the dollar.
Just wondering, why would you choose nominal over real?
Posted by: Buzzcut at April 15, 2008 01:04 PM
1) A tax on imported oil, to reduce US consumption while forcing importers to shoulder some of the burden.
With the economy weakening and the price of crude surging, is that really necessary?
As JDH noted, light truck sales were less than car sales for the first time in a looooooong time. Gasoline demand fell for the first time in a long time as well.
Americans are changing their habits, clearly. If demand for gasoline falls 1%, demand for imported oil should fall almost as far.
Posted by: Buzzcut at April 15, 2008 01:08 PM
Curious that there was no mention here of the last coordinated intervention in the forex markets, which arguably "worked." That was at the last peak of the dollar/trough of the euro, during its introductory transitional phase in September 2000. It is my understanding that Larry Summers went along with demands from Tokyo and Frankfurt/Paris to join in a coordinated intervention to prop up the euro. It was brief, but it did mark the turning point, if I am not mistaken.
BTW, Menzie, are you still so sure that the long swings that Jim Hamilton and Charlie Engel once saw in the dollar's movements aren't really there and that the random walk is more realistic?
Posted by: Barkley Rosser at April 15, 2008 01:28 PM
But a tax on imported oil will cause some of the burden to be shouldered by countries we import from, AND will accelerate the shifts in US behavior.
For a change, Saudi Arabia, Venezuela, etc. will be financing a shift in US consumer behavior away from oil, whether they like it or not.
If ever there was one instance where a tariff on imports would be good, it would be on imported oil.
Posted by: GK at April 15, 2008 01:28 PM
Thanks for nudging me to take a second look at the graph.
It is interesting to note that from the time Nixon took us off of gold until the 1980s there was not much fluctuation of international exchange rates, but as those of us who lived through it know, this was the period of the Great Inflation as all currencies fell in value together, worldwide inflation in all currencies at about the same rate.
This makes sense for two reasons. First, after Bretton Woods the world was not on a gold standard but on a gold exchange standard. The difference is that many smaller countries held dollars in reserve rather than gold, assuming dollars woudl be a proxy or "exchange" for gold. When Nixon took the US off of the gold standard these countries inflated with the dollar because that was their reserve. Secondly, most monetary regimes were linked to the dollar rather than gold (again Bretton Woods) and so followed the dollar into chronic (some of them hyper) inflation.
Note that with the institution of the Supply Side policies of the Reagan administration during the 1980s the dollar strengthened against all other currencies. US prosperity was pulling the world out of Nixon's inflation.
But the strong dollar became a concern for primarily academic reasons, so in 1985 the Plaza Accord agreement engineered the decline in the value of the dollar. Because this was an artificial debasement of the dollar the US economy began to waiver and the decline ran out of control. In 1987 came the Louvre Accord to try to correct the problem caused by the Plaza Accord.
Then on August 11, 1987 Greenspan was confirmed as FED Chairman and on Sept 4, one of his first actions, Greenspan raised the discount rate from 5 1/2% to 6% shaking the markets. On Oct. 18 Treasury Secretary James Baker stated that the dollar should continue to fall and threatened trade actions against Germany and Japan because of currency issues. The next day the market had the largest percentage crash in its history.
Then during the 1990s after a short flirt with higher taxes which gave the House and Senate to the Republicans, Bill Clinton, very possibly without knowing it, began implementing supply side policies. All currencies stabilized during this time trade boomed as did world economies.
Then came the Greenspan deflation of the late 1990s when he determined that there could be no deflation with a floating dollar. The deflation lasted until Greenspan saw retirement rushing at him and began to think of his legacy. He promptly reversed his deflationary stance and began a rapid inflation so that when he retired the economy would be in the midst of an inflationary boom. What he didn't consider was that he might simply create a credit boom which, after he retired and Bernanke began to rein in the inflation, turned into a credit bust.
All of this can be seen in this graph.
Posted by: DickF at April 15, 2008 03:41 PM
GK: Raising the tariff on oil is not, I think, WTO consistent. Note that a tariff on oil would encourage more domestic production, which is also the same as more rapid depletion. In any case a tax on oil would be welfare-improving if the distortion is a negative externality associated with oil consumption.
Faster yuan appreciation would help reduce the US-China trade deficit, but would mostly re-allocate it to other countries unless other East Asian currencies move in line. See this paper.
Buzzcut: I wanted March's data in the plot, and for the real series, March's CPIs are largely estimated. Hence, selection of the nominal series. The movements in the major currencies indices -- real vs. nominal -- do not differ much. There's a trend, but there's a trend in the CPI-deflated series as well (and should be if there's a Balassa-Samuelson effect).
Barkley Rosser: I was trying to summarize the academic literature on intervention, and I think I gave a balanced summary. I allow that sterilized forex intervention might have an effect on the level (see the references for the Japanese case). In addition, signalling could be important. The signalling channel was discussed by Frankel and Dominguez, among others.
Regarding the long swings view, I'm not dogmatic on the issue. I do think changes in exchange rates can be modeled as a Markov switching process. However, Engel (JIE, 1994) showed that a switching process did not outperform a random walk characterization in out of sample forecasting. And that's all I claim. I actually spend most of my time trying to outpredict a random walk only because that's the convention in the literature.
Posted by: Menzie Chinn at April 15, 2008 08:14 PM
While I'm sure the euro is destined for a reserve currency status, I have a hard time understanding the degree of breathlessness to which it is discussed. Two major reserve currencies within several percentage points of each other in terms of composition of exchange reserves seems remarkably different to me than one currency comprising two to three times those reserves as another.
Posted by: mm at April 16, 2008 11:47 AM
A reserve currency is normally thought of as a currency that will be a store of value against other currencies. The US$ has fallen against the euro from 1$/1€ to 1.5$/1€. Not what I would call stable.
Posted by: DickF at April 16, 2008 02:45 PM
These are certainly exciting times for currency speculators. My guess - the euro will shortly be looking much less attractive. While it's true that the ECB has greater inflation discipline, I believe this is because they have more rigid labor markets, so inflationary expectations are harder to deal with once they take hold. This may mean that they are more susceptible to deeper recessions, though.
People in the euro area appear to have more saving discipline than Americans and there are already huge stocks of U.S. dollars in foreign reserves, but I think this is a longer run problem.
Why do you cite the change in the $/euro rate as showing instability of the dollar? Why doesn't it imply instability of the euro?
Posted by: don at April 16, 2008 09:04 PM
Another very ueful and informative post, but I think it would be even better if some notion of purchasing power parity could be displayed, even though it would be subject to great error. For example, it would be nice to know if 4.5 or 4.6 were closer to the parity value over this period.
Posted by: don at April 16, 2008 09:13 PM
Don: The problem is -- who knows what the equilibrium PPP exchange rate for the dollar is? I discuss this issue in this post (where you might find the graphs you're interested in). Note however, these graphs rely upon relative PPP. For an explanation of the distinction between absolute and relative PPP, see this set of notes.
Posted by: Menzie Chinn at April 17, 2008 08:49 AM
The IMF apparently concluded not too long ago that the dollar was still 10%-30% overvalued when the euro was at $1.38. I wonder if they accounted properly for the VAT? I know The Economist's Big Mac index does not.
Posted by: don at April 17, 2008 08:08 PM