Myths of the Adjustment Process
Woody Brock’s February 2004 paper on the relationship between US interest rates and the big dollar is a neat debunking of much of the conventional wisdom on this subject:
• If foreigners become disenchanted with US assets, and demand higher yields, can they get them? [No, other things being equal.]
• Can foreigners as a whole “pull out” their money from the US, thus driving up US real rates? [No.]
• Can foreigners as a whole refuse to acquire more US assets in the future (in order to finance future US trade deficits) thus driving up US interest rates? [No.]
• As the dollar falls, will the US experience a dose of “imported inflation”, thus driving up nominal interest rates? [Yes, but less than ever before due to structural changes in the global economy.]
• Can Asian central banks stop acquiring US IOUs? [Yes.] And if so, would this development send US interest rates soaring as the consensus expects? [No, the value of the dollar would take the hit much more than US yields.]
• How much more will the dollar fall before a new and more stable equilibrium can be achieved - an equilibrium including a balanced US trade account? [Much more than it has - largely because the value of the dollar has not been the cause of today’s trade imbalances.]
Woody greatly overstates the likely exchange rate adjustment in my view (eg, AUD-USD above parity,* a 300% revaluation in the yuan), but his analysis of the irrelevance of this process to the determination of US interest rates is essentially correct.
* Afterall, Australia has its own record current account deficit, for which a similar adjustment process is required.
posted on 26 November 2005 by skirchner
in Economics
(4) Comments | Permalink | Main
|
Comments
Stephen—frankly, many of the points brock makes, in there strongest form, are ridiculous. He takes points that are true—foreigners hold $3 trillion (now closer to four) in claims on the US economy, and cannot reduce their holdings in aggregate unless the US runs a current account surplus (true), and then concludes that foreigners willingness to fund the US deficit has no impact on US rates (a point refuted by the WArnock study you have cited before).
Think of it this way—foreigners have to hold $4 trillion in (net) claims on the US in equilibrium. And they have to add $800b plus to their net claims to fund ongoing US deficits. (Yes, by accounting, any deficit has to be funded, but it the int. rate that matches supply and demand can be rather different). But if foreigners expected the dollar to depreciate, they would tend to sell those claims—and while their face value would still be $4 trillion, their market value would fall (raising interest rates). Someone else would eventually buy the claim, but at a lower price (and thus with a higher yield if it is a bond). And on a flow basis, if foreigners are unwilling to increase their holdings of claims on the US at current rates, rates would need to raise—that would equilbriate supply at demand, because with higher rates, more folks abroad would be willing to finance the US, and there would be less demand for funds in the US. But rates would still increase.
Look at what happened to interest rates in Argentina when foreign investors lost confidence in Argentina—they went up. Way up. the market value of the existing stock of Argentine eurobonds fell. And Argentina had to promise to pay a higher interest rate to continue to attract the new flows required to keep on running its current account deficit (yes, the total stock of dollar claims—at face—still went up).
This really isn’t hard.
Nothing rules out a similar adjustment process in the US. It may not happen, granted. Other adjustment paths with low interest rates are possible. As is a continued period of no adjustment. But it could.
the key difference—claims in the US are denominated in the United States own currency.
One point of agreement. If the dollar falls massively, then interest rates could stay low—because foreign investors would expect a future appreciation. But the key is having a big enough fall to generate expectations of future appreciation ...
Posted by .(JavaScript must be enabled to view this email address) on 11/28 at 03:12 AM
Actually, I think the Warnocks probably overstate the implications for US interest rates, but I’m willing to run with their empirical work for the sake of argument and in the absence of better estimates.
“they would tend to sell those claims—and while their face value would still be $4 trillion, their market value would fall (raising interest rates). Someone else would eventually buy the claim, but at a lower price (and thus with a higher yield if it is a bond).” This doesn’t necessarily follow. Financial market prices are not just determined by flows, especially on the part of just one part of the market for these assets.
I think comparisons with Argentina are a complete stretch for the US, for reasons you hint at.
The sort of USD depreciation Brock talks about I would have thought was something that ties in with your own views, except that Brock takes a benign view of this adjustment, whereas you do not.
Posted by skirchner on 11/28 at 11:23 AM
I don’t quite follow how you argue “selling = lower prices = higher yields” doesn’t follow. seems pretty simple. Markets clear—and but prices adjust. and when financial market prices adjust, that usually involves some change in yield. Claims on the uS fall in price until the higher yield induces someone to buy ...
Both stocks and flows matter—and in that respect, Argentina is like the US.
Argentina had something like $160 billion in extenral debt and needed to add $10 billion to that stock a year to finance its current account deficit. So after a year, in equilibrium, it needed to find holders of $170 billion of its debt (numbers approximate). Because of arbitrage, the yield it needed to offer to get the incremental $10b in new financing helped set the market clearing price for the remaining $160b. At the end of year 1999, it might have had $160b in debt outstanding (face) with roughly $150b in the market. At the end of 2000, it had maybe $170b outstanding, worth say $155b ... (numbers very approximate). by the end of 2001, it had $170b outstanding, worth maybe $40b-50b ...
Here is how that relates to the uS: the US has net external debt of roughly $4 trillion right now. Barring valuation changes, that will need to increase to $4.9 trillion next year. And i suspect the price/ yield needed to attract the $900 b in net new financing will help determine the market value (price/ yield) of the outstanding stock as well.
Posted by .(JavaScript must be enabled to view this email address) on 11/29 at 11:00 AM
“Afterall, Australia has its own record current account deficit, for which a similar adjustment process is required.”
...and it ain’t getting any better…
Growth in exports hits a 50-year low
http://www.theaustralian.news.com.au/common/story_page/0,5744,17397924%255E2702,00.html
Posted by .(JavaScript must be enabled to view this email address) on 11/29 at 02:33 PM