I'm sick of the dollar losing value!

[DanS]

That chart appears to contain errors. F.e., the 1997 average monthly trade-weighted exchange rate is 104.43 rather than ~ 85 in that chart. In 1973, it is 31.70 rather than 100, so its not a matter of having a different index year.

http://research.stlouisfed.org/fred2/series/TWEXBMTH?cid=105

Not sure what the "real" part of the exchange rate that is referred to in that chart.

Who prepared the chart?

[Colon™]

I didn't realise the current account deficit is only 0.06% of GDP.

[DanS]

If MZ prepared the chart on his spare time, I would be inclined to forgive the formatting stuff.

But I'm still not convinced that a consideration of inflation has any place in this chart.

[Lancer]

44.8

44.8!!!

[Lancer]

How much of this is due to Iraq?

[DanS]

Almost nothing. Mostly, Uncle Sam buys American. Unlike you.

[Master Zen]

The point is moot: your graph (which I take it is different because it is in nominal rather than real terms) still doesn't align particularly well with the current account deficit if what you're trying to prove is an undeniable empirical relationship between the two.

[Colon™]

Uhm, even in your graph, I'm spotting a lagged relationship, be it not a 1-to-1.

[Lancer]

It's not my fault!!!

[Colon™]

Originally posted by Colon™
Uhm, even in your graph, I'm spotting a lagged relationship, be it not a 1-to-1.

Moreover, the time period is too short, the only disproving the relationship would be the later 90's and sample-range too limited, in not considering the experiences of other countries.

[DanS]

Originally posted by Master Zen
The point is moot: your graph (which I take it is different because it is in nominal rather than real terms) still doesn't align particularly well with the current account deficit if what you're trying to prove is an undeniable empirical relationship between the two.

Who decided to put a consideration of inflation into this chart? Was it you?

[pchang]

what exactly is the trade deficit?

What does it really mean anymore? I'm not sure it really includes enough things to be usefule anymore. By some measurements, everything has to balance out.

[Master Zen]

Originally posted by DanS
Who decided to put a consideration of inflation into this chart? Was it you?

*cough* straw-man *cough

If we use my graph, the real data doesn't match. If we use your graph, the nominal data doesn't match either. Conclusion: exchange rates and deficits are not as correlated as the pundits claim. Second conclusion: "Buy American" won't do that much to make the dollar stronger.

That, btw, was my point all along.

[Master Zen]

Originally posted by Colon™
Moreover, the time period is too short, the only disproving the relationship would be the later 90's and sample-range too limited, in not considering the experiences of other countries.

The sample range is enough, trust me. There's really no point in using pre-1973 data since exchange rates back then were largely based on the Bretton Woods standards. Post 1973 is the only continuous recent time period in which data on floating currencies can be observed.

Regarding other countries, the relationship between deficit and exchange rates will be much more observable than the US and would in fact be far more likely to fit the textbook model of a relationship between the two (although even this is open to doubt). For example, one of the reasons why the US is somewhat insulated from these effects is because most US trade is handled in dollars, regardless of who the exporter is (a unique situation for the US since the dollar is the most widely used currency in the world).

Most importantly and this is something that as an economist I cannot stress enough: THERE ARE NO DOGMAS. Empirical phenomena frequently throws textbook theories to the garbage bin. Explaining a statistic as complex by merely considering the exchange rate is not enough: there are many more things to consider.

The first and foremost consideration which DanS is blatantly ignoring is that the dollar has not depreciated versus all currencies. It is quite strong to the peso and quite strong to the yuan. Guess what: Mexico and China happen to be the two of the US's top four major trading partners. Limiting an explanation of the exchange rate only to, say, Europe or Japan, is creating an incomplete picture of the issue.

Deficit reduction policies affect many indicators such as inflation, public demand, etc. Depending on which are actually affected can you actually determine whether it will have a positive or a negative effect on exchange rates. The size of the US economy, coupled with its trade links to the world make it even more foolish to say that two indicators explain each other almost exclusively.

[Colon™]

There indeed isn't a point in using Bretton Wood era data but that doesn't mean that the time period is long enough to draw firm conclusion about the relationship (the nature of which I am not commenting on BTW).
Also, your assumption is that if the relationship isn't immediate and 1-to-1 it doesn't exist. If you'd draw a graph comparing the fed's funds rate with GDP growth there wouldn't be an apparent relationship either, unless you consider lag and a variable impact.