It is purely because of transactional reasons. There is no reason to view the dollar as any safer than the yen, for example, as in Japan their banks are in pristine condition.

The simple explanation is to assume the two flows have no causal relation to each other.
The trade deficit was produced or due to the purchases of foreign made goods in every nation in the world that trades with the U.S. The outflows were produced by the market forces for currency and currency transformation into other forms of financial assets.

Distaste for the barter system is the only one I can think of.

Valuation is of course the whole point. But why the valuations? And how does the change in perceptions show up in flows, or perhaps short term dislocations in dollar denominated markets is what is good to try to get a handle on.

The yuan is not anything like a freely traded currency. But in the case of the rest of the BRICs, FDI in hot stock markets, and maybe in direct bank lending, had a lot to do with currency valuations. On the order of trade factors at least. It also swelled apparent CB reserves in these countries.

Also, the Euro is a new major currency challenging the dollar as a reserve currency, or at least being a candidate for a diversified basket of currencies, which is one solution to global currency risk.

In the case of the yen, Japan has a substantial amount of private investment. ZIRP and quantitative easing caused all this to go looking elsewhere for yield, mostly to the US. The unwinding of this is a major factor in yen/dollar strength.

So FDI flows are important. In the case of the ’90s dollar, we first had Greenspan rapidly raising interest rates in 1994, giving the dollar a boost. We also had relatively modest twin deficits. But then the late ’90s stock market boom attracted a lot of FDI chasing it and the dollar index hit near 120 ! Far cry from today. Now it’s looking like a Treasury boom could take it as high as 85.

Then the futures markets decide on what they think the future spreads are between cross currency interest rates.

So a massive amount of economic analysis analysis and forecasting happens. Then they do over again the next day with any new news or data that comes out.

I’ve had similar thoughts of heresy. Generally I try to limit my thinking to FX transactions (or lack of them in the case of surplus CBs re-cycling dollars to peg currencies) are the cause of currency movements. I’ll even believe interest rate futures markets in currencies have a lot to do with it, since a lot of trade goods and financial instruments are legitimately currency hedged this way. And I’ve seen correlation charts to prove it, tho like you say its hard to tell from a chart who’s chasing whom.

And I think a trade deficit puts pressure on FX if it doesn’t all come back in re-cycled dollars.

And I generally believe that if you use dollars to buy and sell oil in dollars, or anything else denominated in dollars, that is currency neutral even if you do it from Grand Cayman. (other that trade deficit effects not offset by re-cycling).

But I do get a nagging suspicion that if enough oil spectulators are involved, like enough to move the oil market between $140 and $70, and they are using huge leverage, and they all start getting margin calls, they all start screaming…I NEED DOLLARS !

So I think that could have an impact over at least a few months.

Really enjoying having found your blog, but I’m not a big fan of the tic data, (or flow of funds analysis wrt exchange rate determination for that matter). Having spent over a decade on the FX desks of 3 different global banks, I’ve come to the conclusion that the FX market discovers fx rates, but it generally does not determine them.

to wit:

Lets assume that the Dollar is stable on foreign exchanges and the both the supply and demand for Dollars and the supply and demand for, lets say, Euro are in perfect equilibrium. In this fantasy world, I am a multi-billionaire, of course. Suddenly, I decide I’m afraid of the banking system and I withdraw $25bn each from my accounts at Citi, JPM, Wells, and BoA, and I bury the money in a gigantic coffee can in my back yard. The equilibrium has now been upset as the result of my spontaneous, massive demand for Dollar liquidity. It’s illogical to think that I haven’t effected the value of the dollar by my actions, just because I haven’t acted directly in the FX markets. I would contend that what would happen next, the equilibrium having been upset, is that the next person who needed to buy USD and sell Euro, for whatever reason, would find the price “gapping” against him towards wherever the new equilibrium might be.

I’m not sure what such a chain of events would show in the TIC data, but its my suspicion that in the current environment cross-border transaction data will prove a poor indicator of FX movements.