September 26, 2005
On the origin of American current account deficits
Here are some more thoughts on the debate over the source of the U.S. current account deficit and whether it matters.
There has been a tremendous amount of debate regarding the origin and implications of the U.S. current account deficit that has developed, especially over the past couple of years. With the deficits showing no sign of abatement-- indeed the consensus is for the deficit to yawn even wider to 7% of GDP in 2006-- now seems a good time to provide a little empirical content, leavened with some theory, to the debate over the origins of the imbalance. This is especially true since there is a marked division between those who believe the ever expanding deficit is cause for alarm, and those who argue that not only is the deficit not worrisome, but it is even a good thing.
Growth and deficits
First, let's examine the argument that current account deficits are either a sign of good times, or an unalloyed good in and of themselves. According to this view, the current account surpluses in Europe and Japan are the result of lackluster economic growth and consumption. Proponents of this view dismiss the idea that the United States should take active steps to shrink the current account deficit, characterizing such actions as proposals to throw the U.S. into recession, so that growth rates here will be as slow as those in Germany or Japan. (They also ignore the fact that even with faster U.S. growth during the 1990's, the deficit was smaller.) Often, they argue that the deficit will be remedied if the rest of the world follows the U.S. lead in implementing pro-market reforms. The eventual growth bonus arising from such policies will turn current account surplus economies-- the euro area, Japan-- into deficit economies. Unfortunately for this cheery worldview, the empirical evidence is not supportive; as Catherine Mann and Katherina Plueck at IIE (Figure 4) have pointed out, even a robust-- but still plausible-- acceleration of growth abroad will do little to reduce the deficit.
A related argument relies on the intertemporal model of current account balances. In this framework, deficits signal future economic strength. Ignoring the lack of empirical evidence for this approach, this argument would be more convincing if U.S. GDP growth were being maintained by investment rather than consumption and, more importantly, if the lending to the United States took the form of purchases of stock and direct investment. Instead, a large proportion of capital flowing to the United States takes place in the form of purchases of U.S. government securities-- not purchases of American stocks or direct investment in its factories. The fact that foreign central banks are doing so much of the lending suggests that the profit motive is not behind the ongoing flows to the United States.
Saved by revaluation effects?
A second argument relies upon the observation that, despite the recent record deficits, the net international investment position did not deteriorate going from 2003 to 2004, and is (only!) -21% of GDP. This phenomenon occurs because most U.S. overseas assets are denominated in foreign currencies, and when the dollar loses value, most of America's assets gain value. But we can only rely upon this effect working in our favor if either of two conditions holds: (1) unanticipated dollar depreciation continues, or (2) foreign investors remain happy with their present holdings of American assets at current interest rates. I believe neither is likely to hold true indefinitely. (Some seem to hope that dollar depreciation continues indefinitely; Jeff Frankel and I argue that this will lead to erosion of the dollar's role as the world's key reserve currency.)
If the future is roughly like the past, then a trade deficit of 1.4 percent of GDP is consistent with stabilization of the level of indebtedness. For sure, this is smaller than the 7% that seems to be the consensus number for 2005. Hence, we are headed toward greater and greater indebtedness.
The "Savings Glut" view
Finally, there is the very popular view, most closely associated with the new CEA Chair Ben Bernanke, that there is a global "savings glut": The large current account surpluses in the rest of the world, particularly in China, but more generally in East Asia and continental Europe, are at the heart of the pattern of global imbalances. These current account surpluses have to be offset somewhere, and that somewhere is in the United States, largely because of the greater attractiveness of American assets.
While the savings glut view has some intellectual merit, particularly for the last couple of years, there is reason to suspect that much of its newfound popularity stems from how it conveniently absolves U.S. elected officials from taking action. After all, in this worldview, fiscal policy cannot really have an impact on the current account deficit.
This is an odd argument. First, over half of the financing of the U.S. current account in 2004 was accounted for by accumulation of dollars and U.S. Treasury securities by foreign central banks. This is not a "savings glut" in the sense of excess private savings flowing to the United States. Second, it does not make sense to think of the East Asians essentially forcing the United States to consume beyond its means and borrow from them. The idea that developments in the United States are driven by those in East Asia appears at variance with common sense when the U.S. economy is three times the size of developing and industrializing East Asia and 30 percent larger even after Japan is combined with this grouping. In addition, my recent work with Hiro Ito finds that out-of-sample estimation for the 2001-03 period indicates little role for special circumstances such as the "savings glut".
What's the other possibility?
I think that the conventional wisdom is more plausible: there is a savings scarcity in the United States, driven largely by the federal budget deficit, and it is this savings drought in the United States that has been sucking in excess savings from the rest of the world for most of the past five years. What about the low interest rates? In my view, the low interest rates are more the function of expansionary monetary policy, and corporate savings at home, rather than excess savings abroad (except for government/central bank purchases of U.S. treasuries). (And let's be clear about where the low U.S. savings is primarily coming from: it is government dis-savings that is doing most of the harm here; blaming low household savings is a smokescreen, since the private savings rate in 2004 was slightly higher than it was five years before that, largely due to a surge in corporate savings.) Hence, notwithstanding Greenspan's assertions, there is probably substantial scope for external adjustment via budget deficit reduction (as in OECD analyses and IMF simulations).
Why we shouldn't be complacent
To the extent that savings abroad are to blame for some portion of the U.S. current account deficit, there are reasons to doubt the durability of continued East Asian lending to the United States. First, Chinese lending to the United States in the form of purchases of U.S. Treasury securities can continue only as long as the People's Bank of China can prevent the expanding reserve accumulations from spilling over into money creation. Otherwise, inflation will accelerate, eroding Chinese competitiveness or spurring capital flight. Second, if East Asian investment rebounds, interest rates will rise at exactly the same time as U.S. national savings is declining.
The consequences of a collision of expanding U.S. budget deficits and rising interest rates abroad is not pleasant. Current account adjustment in the United States will entail adjustment of factors of production out of the nontradables sector (housing, some services) into the tradables sector. Interest rates in the U.S. would rise relative to what they would otherwise be. Interest sensitive sectors will take an additional hit. Even if a scenario of financial disruption (say in the hedge funds or in other derivative markets don’t take a hit) is avoided, there could still be substantial economic pain – pain that might be avoidable if action is taken sooner rather than later.
Posted by Menzie Chinn at September 26, 2005 04:31 PMdigg this | reddit
Let's assume that the Federal Government drastically reduces its deficit, and lack of savings, by raising taxes 400 B$/year. Consumers and companies spend about 15% of their income on foreign goods. If this percentage stays constant, then the purchase of foreign goods would be reduced by 60 B$, which would not make a significant dent in the Current Account Deficit. So, I don't understand why you associate the trade deficit with a lack of government savings.
Posted by: SteveB at September 26, 2005 06:13 PM
Nice piece Menzie, and welcome!
Posted by: RH at September 26, 2005 08:03 PM
There is an ongoing debate regarding the size of the effect going from the budget balance to the current account. Greenspan, citing research at the Fed (the Sigma model), argues that the effect is about 20 cents on the dollar -- comparable to the effect you cite. However, other research groups have obtained higher estimates. For instance, as indicated in OECD Working Paper 306, the effect is about 40 cents on the dollar.
The typical rebuttal is that new generation (Dynamic Stochastic General Equilibrium, DSGE) models like the Fed's Sigma model are more reliable. However, the IMF's Global Economic Model (another DSGE) yields a 50 cents on the dollar effect.
So, I have to admit, I don't understand why so many people fail to see an important effect. If, for instance, we have a fiscal consolidation like that experienced during the 1990's, then the current account deficit would improve by up to 3 percentage points of GDP.
Posted by: menzie chinn at September 26, 2005 08:03 PM
You and Steve Roach are singing from the same hymnal. You guys could became a barbershop duet.
It will be very intesting to see who is right as this plays out over the next decade.
I am hoping that the smart guys at the fed have some super mojo that will prevent the hard landing.
Posted by: dilbert dogbert at September 26, 2005 08:46 PM
Professor, welcome and thanks for the great post!
I suspect the impact of the fiscal deficit is close to $0.40 on the dollar. I also think that the emergency monetary policy over the last 4+ years has contributed signficantly to the CAD.
Greg Ip has a nice article in the WSJ tomorrow: Greenspan Warns of Reliance On Housing Loans
He discusses Greenspan's recent paper / comments regarding mortgage equity extraction and the contribution to consumer spending. So the housing boom, especially the financing aspects of the boom, has probably contributed to the CAD too.
Once again - Welcome. I hope you are a regular contributor to Dr. Hamilton's excellent blog.
Posted by: CalculatedRisk at September 26, 2005 09:07 PM
Thanks for the great feedback. To Dilbert Dogbert, well, I don't know if Steve Roach and I really see eye to eye. I am not forecasting a hard landing; all I'm saying is that regardless of the source of the deficit, and end to the current account deficit may be more difficult than is portrayed in various rosy scenarios. Suppose it is a savings glut driving the U.S. current account deficit. Suppose further the excess savings flowing from East Asia is cyclical, not secular, in nature. Then a abrupt cessation of capital flows from East Asia and the oil exporting countries will force rapid adjustment of either the dollar, interest rates, or output. Since the repercussion effects are likely to be large (remember the saying, "When the U.S. sneezes, the rest of the world catches the cold"?), I think it pays to err on the side of caution.
If I'm right I'm certain the folks at the Fed will do a good job; I know many of the staff and even a few on the FOMC. However, if there's anything economics teaches you, it's that one is confronted with trade-offs, and sometimes the policy frontier you face is just plain unpleasant.
To Calculated Risk, thanks for the supportive words. I haven't read the WSJ article (no access, I'm afraid). But I do agree that the booms in the equity, and then subsequently real estate markets, have probably had something to do with our CA deficit. Hiro Ito and I are just now finalizing a paper where we find that a standard model of the current account balance (as in Chinn & Prasad, JIE 2003) underpredicts the deficit by 1.7 percentage points; these prediction errors are highly correlated with movements in the equity index.
Posted by: menzie chinn at September 26, 2005 09:29 PM
Really a relevant post, thanks! But I would like to wonder why we couldn't see the Asian-US imbalance as the result of interaction of the US budget deficit and the Asian savings glut. The US deficit couldn't be that high without the Asians financing it from their savings. The US government couldn't sell the treasuries at such a low rate without the Asian demand.
Besides I think that we cannot look the world as US-centric as before. China alone is now the biggest exporter in world (it doesn't show in the statistics yet but at this growth rate it has probably happened by now). It is not only the GDP but also foreign trade that gives leverage in the world economy.
I see the main root cause of the present imbalances in the huge structural change in investment and saving. We might say that the bulk of productive investments has moved from OECD to China (and India, Indonesia, Brazil). This shows in the fact that the aggregate industrial production of the G8-countries is now smaller than the new industralized countries (China is now the leading industrial country in the world).
We could say that the US is kept floating by the most robust and dynamic economy in the world - the Chinese. China keeps the world economy growing and drags everybody along. In this context there will be no hard landing before the Chinese growth slows markedly.
But then there is a new factor, the oil price. The oil producing countries are just now having a huge increase in their surpluses which affects the US-Asian imbalance system. The flows of the oil money is not yet clearly seen but it is probable that they will increase the imbalances and make the situation more unstable. I would like to see some comments on this.
Posted by: TI at September 27, 2005 01:21 AM
Without agreeing or disagreeing, it seems to me that a key issue is that much of the funding of the US deficit is coming from central banks. And, as you rightly point out, these central banks' dollar purchases are not profit motivated. They are resisting currency appreciation. And they seem likely to continue to do so until growth in their economies looks overheated. So let me play devil's advocate.
Suppose the federal government cuts the budget deficit by 2% of GDP (for starters). This likely depresses US growth somewhat, at least initially. That makes Asian central banks even more nervous about life because export share in GDP is today higher than ever before so weaker US growth hurts them badly. They respond by buying more dollars to keep their currencies cheap if not try to cheapen them further to repair exports. The combined impact of the recycling of these dollars and the lower deficit in the US depress treasury yields and lead to further consumer dissaving via lower mortgage yields...or something like that.
For the sake of argument, the point here is to ask what the US can do if the world pegs to the USD and in the process forces capital flows into the US? Alternatively, why do you think it is not credible to believe that the Dooley, Faulkers, Landau new pegged rates regime can unwind slowly via inflation and real exchange rate appreciation in Asia rather than via US dollar depreciation?
Posted by: RF at September 27, 2005 03:18 AM
To clarify, I am wondering if as long as EM central banks peg, all that happens with US fiscal consolidation is that you shift the sources of the US current account deficit from the government sector to the private sector without changing the level as a % of GDP.
Posted by: RF at September 27, 2005 03:22 AM
So we see that the fate of the imbalances is in the Asian hands. What could the US do here?
The real question is what would the appreciation of the Asian currencies really mean. In the context of rising oil prices the appreciation could mean making the import inputs cheaper and keeping up domestic growth.
In any case the oil prices complicate the US-Asian relationship. There arises a need to finance not only the import of Chinese goods but also oil import. How much of the oil dollars will be recycled back to the US? How much this would increase the indebtness and deficit in the US? What will China and Japan do if the US tries to curb the increasing deficits? How much will the rising costs of oil import cut from the trade surplus of Japan?
Posted by: TI at September 27, 2005 04:02 AM
Some excellent insights from TI and RF. Let me try to break them up into pieces. (I'll have to punt on the issue of recycling of oil dollars -- I haven't thought about the quantitative magnitude of this effect, but it could be significant, like a percentage point of the deficit to GDP ratio.)
Is the U.S. budget deficit endogenous, and reacting to the availability of foreign savings. In typical macroeconomic modeling, such assumptions are avoided. But, departing from the models, I guess it is plausible that the easy availability of credit to the U.S. government in the form of East Asian central bank willingness to purchase U.S. Treasuries may have "greased the skids" for our "comfortable path to ruin" (sadly, Martin Wolf's turn of phrase, not mine). That is, the interest component of the Federal budget has been kept relatively small by historically low interest rates (when they rise, watch out!).
Turning to how this links to the exogeneity (or lack of) of the savings glut -- I think it's interesting to ask the hypothetical: if the borrowing needs had come from American corporations like Enron, would the East Asians have been so willing to lend. That points out the hazards of combining accounting identities and assumptions of exogeneity.
Another important issue is the durability of the East Asian pegs and the "excess savings" flow. If Dooley et al. is correct, then it'll be hard to do anything, from a policy perspective. However, the mercantilist interpretation faces some difficulties. Why did they start "undervaluing" their currencies in 1999 (and later for China). Inspection of the current account data for China suggests that the accumulation of reserves is China has been mostly accounted for by capital inflows, not large current account imbalances, over the last couple of years (see Eswar Prasad and Shang-Jin Wei's IMF Working Paper 06/79, http://www.imf.org/external/pubs/cat/longres.cfm?sk=18138.0 ). Also Aizenman and Lee (2005) provide convincing evidence that while statistically significant, mercantilist factors are trumped by precautionary motives in the determination of forex reserve accumulation
( http://www.frbsf.org/economics/conferences/0509/paper-aizenman.pdf )
Finally, will the end of central bank reserve accumulation mean the end of the excess savings glut? I think it will, at least in part. The decrease in investment in EAst Asia ex-China is strongly linked to the financial crisis in the region. This suggests that there is a large transitory component to this. In addition, in China, the surge in savings is largely accounted for by corporate savings (see the box in the latest IMF WEO). These are forecast to shrink over the next year. So once again, I look for an alteration of the patterns.
Posted by: menzie chinn at September 27, 2005 07:21 AM
SteveB: You are considering only the direct effect of reducing the fiscal deficit. The indirect effect is that the Fed would reduce interest rates (or stop increasing them) and thereby weaken the dollar, which would make US goods more attractive relative to foreign goods, thereby further reducing the trade deficit. (I presume this is approximately what goes on in models like the IMF model that Menzie Chinn cited in the response to your comment.)
CR: "the emergency monetary policy over the last 4+ years has contributed significantly to the CAD". This may be sort of true, but I think it is a misleading statement. Any policy that successfully stimulated the economy would have raised the trade deficit. The alternative was to have an ongoing (probably contained) depression, which would have included a lower trade deficit. An alternative form of stimulus (maybe a fiscal stimulus even more aggressive than the one we got) might have produced a recovery, but the effects on the trade deficit would have been even worse than the ones produced by monetary policy (which had the partially offsetting effect weakening the dollar). Among mechanisms that could produce recovery, the monetary response was probably the least damaging to the current account.
Posted by: knzn at September 27, 2005 08:13 AM
Interesting post, and two questions.
First, what about oil prices? Demand for oil is price-inelastic in the short run, and U.S. domestic production is falling even as income-driven demand is rising. Higher oil prices will boost dollar oil imports. Does that mean the U.S. CAD goes up, or do higher oil imports just crowd out other types of imports?
Second, most oil producer/exporters are not experiencing an investment boom, even as cash is roaring in. (Iraq ought to be able to finance reconstruction if it ever attains peace.) Thus, oil countries have increased their saving rate and non-U.S. saving rate. They finance increased U.S. CAD by accumulating the much safer U.S. assets, and push down real interest rates. Bernanke confirmed! This is why the bond market rallies every time oil prices rise, the opposite of the pattern of the 1980s.
Implication: if oil prices are a bubble which bursts, and we end up with $30/bbl oil, there will be a global investment boom and surging real GDP growth, less global saving, higher real interest rates, a lower U.S. CAD, and a somewhat lower U.S. budget deficit.
Posted by: Brian Horrigan at September 27, 2005 08:55 AM
Regarding the issue of monetary policy, it could go either way. Most macroeconometric models note that expansionary monetary policy leads to both expenditure increasing and switching effects. The lower interest rates induce greater aggregate demand and hence elevated imports; but the lower interest rates also depreciate the currency leading to increases in exports and decreases in imports. The countervailing effects dominate at different horizons, with the income effect dominating at the short horizon.
These models are typically based upon historical experience, but there is reason to wonder whether the most recent episode -- wherein the Fed Funds rate was driven to very low levels for a prolonged period -- will result in a different pattern of effects. Along these lines, Ted Truman at IIE has stressed the expansionary (and hence import inducing) effects over the expenditure switching effects. I guess it depends on whether you think there is, or isn't, a monetary policy induced bubble in the housing market.
Regarding petrodollar recycling, it is true that higher oil prices mean that the current account balances of oil exporters are increasing. This should have an immediate effect upon their demand for dollar denominated assets. However, nothing requires them to hold the accumulated reserves in dollar denominated securities, or to buy American physical assets. So, the effect Brian Horrigan mentions holds true to the extent that there is tremendous inertia in how OPEC country central banks accumulate reserve currencies (an open question in my mind).
Posted by: menzie chinn at September 27, 2005 09:26 AM
Three remarks on this interesting post (welcome from a UW-Madison grad).
1) I think your argument that piling up of reserves by Asian central banks will be inflationary is weak. Japan has the world's largest foreign reserves, still climbing, and has been battling deflationary pressures for years. While the China has been inflationary in the past, possible deflation has been the bigger threat more recently. The third largest foreign exchange reserves country, Taiwan, also has a very low inflation rate. It would seem that these countries are very able to sterilize their inflows, indeed might wish they had more of an inflationary effect.
2) Reports I hear out of Asian central banks are that China is now following the Singapore model of a "banded basket crawl." Although the weights have not been publicized it is widely thought they are probably trade-weighted. Given that China is not running much of an aggregate current account surplus, despite its large (and growing) bilateral one with the US, there is no reason to expect much of a general appreciation of the yuan/renmimbi, although perhaps more so with respect to the dollar.
This and the previous point suggest the Asian central banks might be able to accumulate dollars for a longer time, which might actually heighten the danger of a hard landing, albeit one further off in the future.
3) The sharp disjuncture between the competing DSGE models (20% vs 50%) is a sharp reminder that there are competing such models with no clear reason to believe one over the other. DSGE is supposedly the "new macro consensus," at least theoretically, but its bottom line output is a great big mess for policymakers I fear.
Posted by: Barkley Rosser at September 27, 2005 09:55 AM
Dr Menzie -
You say, "pain that might be avoidable if action is taken sooner rather than later."
What action do you recommend ?
Is it ?
a. Stop Interest rate hike ?
b. Curb Chinese exports by quotas, pressure etc ?
surely, you can't simply say "CONSUME LESS".
I feel there is credence to Savings Glut. Its not just the savings of china..its their act of buying our treasuries in loads and keeping our dollar high.
The only solution for all this seems to be the dollar to tank, which the chinese are not going to happen until sometime. (Roach predicts 18 months time from now :-))
Do you have any other "actions" we can take ?
Posted by: Humid at September 27, 2005 10:37 AM
You say "Implication: if oil prices are a bubble which bursts, and we end up with $30/bbl oil, there will be a global investment boom and surging real GDP growth, less global saving"
30$ /bbl == LESS GLOBAL SAVING ?
I disagree. May be less % saving but actual saving will be even more I guess.
With all the profits, chinese will go crazy buying more bills and pushing their GDP growth to massive levels.
Posted by: Humid at September 27, 2005 10:50 AM
Regarding whether monetary policy has contributed to the trade deficit: I think it depends on what you use as your baseline. If the baseline is no monetary stimulus, with all other exogenous forces the same, then the expenditure increasing effect has probably dominated the switching effect, and monetary policy has increased the trade deficit, but it has also prevented a depression. If the baseline is full employment, however, then there is no expenditure increasing effect (since expenditure is always whatever would produce full employment), so monetary policy has unambiguously reduced the trade deficit (compared to whatever else might have produced full employment). Rather than saying that monetary policy has increased the trade deficit, say that the economic recovery has increased the trade deficit, and that monetary policy is what caused the recovery.
Posted by: knzn at September 27, 2005 11:18 AM
Well, it's tough to keep up with all the interesting comments. Thanks to all for the insights. Let me deal first with Barkley's remarks.
1), 2). Will reserve accumulation be inflationary. So far, it hasn't in China -- partly because the quasi-fiscal costs of sterilization hasn't been particularly high (see Brad Setser's CESifo paper on this point, http://www.cesifo-group.de/link/Setzer-uce05-s.pdf ). On the other hand, this is dependent upon interest rates remaining low; if that changes, sterilization costs could jump. In addition, the stock of nonperforming loans in China constitute an enormous contingent liability. If the government has to take them on, the forex reserves constitute a ready resource and may be (already have been) used to recapitalize the banks. That means they may be monetized in the future, and as the forex reserves get bigger, the temptation becomes greater. I agree, adjustment may not occur soon, but when it does, it may be disruptive.
3) The disagreement on results of DSGE models is part modeling differences (Sigma has long run Ricardian agents with short run rule of thumb consumers; IMF's model has an explicit link between govt debt and net foreign debt via an implicit overlapping generations formulation), and partly due to the differing nature of the simulations. Since these models are calibrated to fit the empirical correlations, I personally prefer to rely upon old fashioned macroeconometric models like the OECD's Interlink (which, as I mentioned, yields a 40 cent on the dollar effect)
Regarding Humid's query, that's the topic of a forthcoming Council of Foreign Relations Special Report ( http://www.cfr.org ) I'm writing. So I shouldn't give away the punch line. However, suffice it to say I believe the view that the U.S. CA deficit is primarily and exogenously foreign driven is wrong, and that U.S. fiscal policy *can* be used to shrink the CA deficit.
On knzn's point that monetary policy has no expenditure reduction effect when output is at potential, I think that makes sense if I view the aggregate supply curve as kinked at full employment (so it looks like a backward "L"). I guess I think of the short run aggregate supply curve as fairly flat, and the long run aggregate supply curve as vertical at full employment. Then monetary policy still has an expenditure increasing effect even at full employment.
The IMF and the OECD were interested in kinked AS curves about 8 years ago; I'm not certain that much came out of that literature.
Posted by: menzie chinn at September 27, 2005 11:43 AM
The question of contingent liability of non-performing loans in China is a very big one and one whose outcome is difficult to forecast. Fear about this was a major reason the PRC so strictly controlled capital flows and long had multiple forex rates. The recent move to loosening the forex rate has coincided with a loosening of controls over forex flows and on participation in various hedge markets by the Chinese banks, who are now rushing into "risk management." It is very unclear where all this may lead, but it could be the sort of sudden adjustment you mention.
OTOH, Japan has also had accumulating foreign reserves in an environment with large non-performing loans with pretty free capital mobility for a long period. This coincided with a severe macro stagnation that many have tied to exactly the effort of the banks to adjust to this without failing (which a few did, and some merged with each other). It appears that Japan may have finally unwound this issue largely. Some argue that the slow decline of the real estate bubble was tied to this, with the banks manipulating the gradual decline as real estate was such an important part of the collateral base. A sudden crash of real estate might well have forced a much more sudden adjustment with many more banks failing. The flip side was the long stagnation.
Posted by: Barkley Rosser at September 27, 2005 01:22 PM
I agree, more or less, about the slope of the AS curve, but I wasn't talking about applying monetary policy to an economy already at full employment. I was talking about comparing an economy with full employment achieved by monetary policy to a baseline economy which is always at full employment. (I assume for convenience that we are at full employment now, although I actually think we are nowhere near, but we are certainly a lot closer than 2 years ago.) Of course the result depends on why the baseline economy is always at full employment: vertical SRAS curve, idealized fiscal policy, investment that rises and falls perfectly with availability of resources, or whatever. But no matter what assumption you make to get the baseline economy constantly at full employment, the conclusion will be that the monetary policy economy has at least as low a trade deficit as the baseline economy.
Posted by: knzn at September 27, 2005 01:36 PM
Nice post, Menzie. Well done.
If we're really discussing the "origin of American current account deficits", shouldn't some attention be devoted to the primary and initial underlying basis for such deficits? Trade practices, for example.
Absent the implications of the successive trade deficits, foreign interests may not have had sufficient incentives to acquire long-term U.S. treasuries in attempts to moderate their domestic currency values. As such, the U.S. could have been paying higher yields to finance any fiscal deficits. Hence, the likelihood that some fiscal deficit spending might have been subdued.
Naturally, the ongoing absence of discussion about the implications of trade policy decisions appears to be wrapped tightly in the quiet notion that the existing lag in any such bilateral relationships and statistical measurements will be overcome eventually. That is yet to proven based on presently employed currency exchange mechanisms.
On the origin of American trade deficits (as a driver for current account deficits), the following trade policy changes deserve appropriate attention when compared to U.S. trade balances:
1. NAFTA took effect on January 1, 1994.
2. GATT was formally signed on April 15, 1994.
3. WTO replaced GATT on January 1, 1995.
4. U.S. trade data:
U.S. International Trade In Goods and Services:
Annual totals, 1960 - present
U.S. Trade in Goods - Balance of Payments (BOP) Basis vs. Census Basis
June 10, 2005
Trade in Goods (Imports, Exports and Trade Balance) with China
It is interesting that we seldom read any analysis of trade policy changes as compared to trade balances at any economic blogs. It is as though there is no interest in the factual presentation of such matters. And therefore no relationship whatsoever.
I call it economic denial.
Posted by: Movie Guy at September 27, 2005 04:09 PM
Barkley, I think the way in which contingent liabilities are revealed differ very much under a capital control ridden peg (China) versus a full-convertible floating rate (Japan). So, yes Japan has had a NPL problem (wrote several memos on the topic while on the staff of the Council of Economic Advisers in fact), but the way it manifested itself was not in a discrete attack.
By the way, a cautionary note about China; right now there's plenty of speculative capital flowing in. But there was a lot of capital flight back in the 1990's, so don't rely on the capital controls contain all disruptions in the Chinese economy.
Regarding Movie Guy's comments, I guess I'm with the crowd that believes that trade policy has only second order effects on *aggregate* external imbalances. Changes in trade policies can reallocate bilateral trade deficits (say, from Mexico to China) but have little effect on the aggregate US trade deficit.
Of course, there are effects of these free trade agreements, but I don't think they're primarily trade related, from the American perspective. For instance, Nafta reduced the likelihood of expropriation of foreign direct investment, and hence spurred capital inflows into Mexico (eventually). In the case of China, WTO accession reduced U.S. tariffs against Chinese goods somewhat, but more importantly precommited China to a path of continued external liberalization of -- for instance -- banking. That probably enhanced direct investment, which then increased Chinese exports. So there is a relationship between the trade barrier reductions, but it's not clear that they increase U.S. deficits overall.
For one interpretation of how decreased trade barriers might induce larger import and export flows, take a look at this:
Posted by: menzie chinn at September 27, 2005 05:38 PM
Figure 3 of the paper that you wrote and cited in the above comment illustrates a simple point. Since 1993, U.S. imports have grown faster than exports by a considerable margin. The gap is widening, not shrinking.
If U.S. trade policy only had a secondary effect as you suggested, then what incentives are driving the growing level of offshoring movements by U.S. corporations? The accommodation of U.S. trade policy is the key. The green light is on.
I agree with your remark concerning the movement of production from Mexico to China. Similar shifts from elsewhere in Asia to China have also occurred. This is readily understood.
The U.S. export consumption lag is the issue. While U.S. exports are increasing, such exports to all seven trading regions on the planet are representing a declining share of exports to such regions, according to the 2005 WTO trade report. The factual data is available for confirmation.
If free trade initiatives are to have net benefit for U.S. economy and its citizens standards of living (a requirement of U.S. trade policy initiatives), then the gap has to narrow considerably. If future changes in exchange rates do not accomplish this goal, then an examination of U.S. trade policy is in order. But the reluctance among economists to undertake that effort will be significant.
Meanwhile, the U.S. export to import gap continues to grow, as do the number of corporations seeking offshoring production opportunities.
Posted by: Movie Guy at September 27, 2005 07:17 PM
It's true that imports are rising more rapidly than exports. What we need to see is whether this phenomenon continues when growth rates abroad match those in the U.S., after adjusting for round-tripping (or vertical specialization, as termed by Kei-Mu Yi, JPE 2003).
In any case, I think it is a mistake to worry about outsourcing per se. Outsourcing (via multinational direct foreign investment abroad, or by contracting out to other firms) is just exploiting comparative advantage by a different means.
Once this is understood, then one should realize that we need to identify the trends in what the U.S. has a comparative advantage in -- and it's in services. Now, a cursory glance at the data would indicate that services are too small to matter. But in fact, the point of the paper I cited was that a lot of the goods trade expansion was round tripping. So what would be the ideal test would be to see what is happening to the trade balance after netting out round-tripped trade in goods, as compared to the services balance (the U.S. has a surplus in that category).
On outsourcing, I highly recommend the following paper by Amiti and Wei:
Posted by: menzie chinn at September 27, 2005 10:22 PM
mr menzie chinn
now that interest rates have increased from 1% to 3.75% can we not expect an increase in savings and a decrease in consumption?this may be more so when rates keep going to 4.25%.will this not make a significant dent in the current account deficit?
Posted by: richard at September 28, 2005 12:54 AM
I checked the Chinese statistics, the latest data is January-May 2005. The trade surplus was 21.17 billion $, yearly value would be 50 billion $. This was achieved by an export growth rate of 34%. This confirms what was said above of the origins of the Chinese currency reserves. But it suggests also that if the Chinese cannot keep up export growth rate of well over 20% the trade surplus would vanish. Or if the imports would grow, mainly because increasing oil prices and volumes. We could expect oil to wipe out most of the surplus quite soon.
But China creates trade surpluses for other countries, mainly Taiwan and Japan. This way the effect is bigger than the Chinese surplus by itself.
This stresses the volatility of situation. The Chinese recycle foreign investments to the US and enable Japanese and Taiwanese to keep their reserves up. The underlying force here is the Chinese economic growth of over 10% a year. China is relatively and absolutely the growth center of the world and draws investements as such. If the US can supply Treasuries as no one else in the world so China can supply real investment opportunities like no one else. The Chinese attraction will not change overnight. If you like to build production capacity wouldn't you choose a country with 94% energy self-sufficiency with energy production growing 10% a year, with inexhaustible labor force, functioning infrastructure, stable government and huge domestic and foreign markets?
On the other hand, the rise of oil prices from 30$ to 60$ a barrel increases the money flow from world oil exports roughly by 600 billion $ a year. This is a lot. Where will all this money go? To the US? To China? And from China?
Posted by: TI at September 28, 2005 01:23 AM
Richard: In theory, you're right, especially over a longer horizon. The increase in short term interest rates have a multitude of channels. In the short term, it tends to appreciate the dollar -- or preventing further decline -- delaying trade balance adjustment.
In terms of effects working in the expected direction, it should also spur private savings, although the elasticity is typically thought to be quite low.
Raising short rate should also push up long term rates; so far that hasn't happened (and that is the "conundrum" that Greenspan referred to). When this happens, it should depress investment and aggregate demand, thereby reducing the trade deficit. It should also cool off the boom in the housing market, reducing consumption and raising savings, as measured by the National Income and Product Accounts. Clearly, neither of these effects have shown up yet.
TI: Yes, Chinese net exports are increasing rapidly, but that is a relatively recent phenomenon. Given that the Renminbi has barely budged, it's obvious that the key determinant in this recent surge is the reduction in output (and hence imports of capital goods and raw material inputs). And before you ask "what reduction", be forewarned that Chinese official output statistics exhibit many interesting characteristics, so I am relying upon the "on the ground" experts reading of the Chinese economy.
TI seems to be asserting that investment should all be occurring in China. If indeed more FDI goes to China (and the Chinese themselves invest more in at home), then this will actually work to the benefit of the global system. Higher investment and aggregate demand at home increases their imports, thereby reducing their trade balance. Personally, I think that Chinese savings will go down -- as I mentioned in my post, the surge in private saving in China has been on the part of corporates, not households, and corporate saving looks like it is on the decline. Investment opportunities may also look less desireable over time as the Chinese economy cools, with the effects of administrative controls taking hold.
The issue of petrodollar recycling is an interesting question. It may be that the OPEC central banks try to use the funds to save in the U.S. But if they are not so concerned with liquidity of the assets, they may purchase assets in Europe or Japan. I admit, that's all an open question.
Posted by: menzie chinn at September 28, 2005 08:13 AM
Germany and Japan manage to maintain a trade surplus with the rest of the world, the U. S. does not. What have the Germans got that the U. S. does not have? A slower growing economy? Higher unemployment rate? Those characteristics help hold down imports but growth in exports is fueling the German trade surplus. Why do U. S. exports grow so slowly?
On the import side, the U. S. problem is that we cannot slow economic growth without facing the threat of currency speculators turning away from the dollar enmasse, making the U. S. citizens less wealthy and ensuring a hard landing.
We are stuck with trying to maintain a rapidly growing economy in order to avoid a hard landing. We must run harder to maintain the status quo.
Posted by: W. Raymond Mills at September 28, 2005 11:43 AM
Menzie, it is an interesting view that the Chinese growth may be slowing down. No surprise though, that level of growth is unsustainable in a longer run. This will likely affect the FDI and also imports. As far as I know Germany and Japan have boosted their exports and trade surplus by exporting capital goods to China. China has really been the engine of global economy, that's right.
And it is clear that extraordinary investment and saving rates in China - nearly 50% they say - are also unsustainable. But who would step in the Chinese shoes and be the investment engine of the world? Those shoes are too big for everyone else. Petrodollars will have an effect but it is unlikely that they will create similar investment boom in an environment of rising energy prices.
Posted by: TI at September 28, 2005 01:21 PM
Hiro Ito and I have just completed a paper in which we determine the "normal" current account balance, given the budget balance, demographic variables, and openness (trade and financial). The results we obtain address both of the points just made by Raymond and TI.
Specifically, we find that Japan and Germany are well predicted, while the US deficit is underpredicted by about 1.5-2 percentage points of GDP. On the other hand, East Asian current account surpluses are underpredicted substantially. So there *is* a saving glut -- it's just not clear it caused, or causes, the U.S. trade deficit. By the way, the prediction error for the U.S. is correlated with the stock market index, suggesting that booms (or bubbles, if you are sympathetic to that view) have some role, and once they are fully deflated, the U.S. current account will adjust somewhat.
Our results also do suggest that the East Asian surpluses are not sustainable, and will eventually shrink.
The paper is at: http://www.ssc.wisc.edu/~mchinn/savingsglut_critiqued.pdf
If you want to see some other views, Charles Engel and I have set up a current account sustainability website at:
Posted by: menzie chinn at September 28, 2005 05:50 PM
Understanding Basic Hydrology in WTO Trade
"It's true that imports are rising more rapidly than exports. What we need to see is whether this phenomenon continues when growth rates abroad match those in the U.S., after adjusting for round-tripping (or vertical specialization, as termed by Kei-Mu Yi, JPE 2003)."
"In any case, I think it is a mistake to worry about outsourcing per se. Outsourcing (via multinational direct foreign investment abroad, or by contracting out to other firms) is just exploiting comparative advantage by a different means."
"Once this is understood, then one should realize that we need to identify the trends in what the U.S. has a comparative advantage in -- and it's in services."
(from a follow on post) "So there *is* a saving glut -- it's just not clear it caused, or causes, the U.S. trade deficit."
I appreciate your fine posts and interaction. Particularly the interaction.
I must say that your views regarding outsourcing (including all forms of offshore production by foreign or domestic producers) and whether the so-called 'savings glut' caused or causes the U.S. trade deficit are not surprising. But such views do not dismiss the current economic situations that the U.S. and other developed industrialized nations face. Nor do they offer adequate explanations in my judgment.
Brad Setser raised the considerations of chapter 2 of the IMF's World Economic Outlook in discussing the global 'savings glut' at his web log along with an article from the Economist.
Link 1: http://www.imf.org/external/pubs/ft/weo/2005/02/pdf/chapter2.pdf
Link 2: http://www.economist.com/displaystory.cfm?story_id=4418328
I have been baffled as to why economists ever expected that there would be a shortage of capital once the WTO accords took effect. The final WTO initiative shift, moving large chunks of production to China from other countries (developing and developed industrial economies), zeroed out the need for increasing levels of capital investment. The result was simple: concentration; large scale production. Backed fully by a concentration of a larger share of FDI.
Similarly, it should be no surprise that corporate savings have overtaken household savings as the main source of private sector savings in most developed industrial nations. No surprise at all.
It is becoming increasingly clear that policy economists do not understand the specifics of the production of goods. They appear to lack a full appreciation of the mechanics of mass production, and effects of labor substitution potential when wage levels represent 1/20th or 1/25th of the wage scale in Western nations.
If one understands scale when discussing or anticipating production of goods in the developing world, the answers are obvious.
For similar reasons, your notion that people should not be concerned about foreign production displacing domestic production and related levels of employee earnings from such domestic production is a bit of a stretch.
The economists who have staked out their ground around present economic models which fail to fully appreciate the fundamentals of hydrology are in a bit of a bind. Explaining simple things like current account deficits in the U.S. or surpluses in China and other developing countries under the operation of the WTO trade regime becomes a difficult task when expected outcomes do not occur.
Under the WTO rules and required compliance requirements by member nations, production of many goods and some services follow the hydrology model.
Free flowing water seeks the lowest ground level before pooling. Once pooled, the additional inbound flow of water seeks another low ground level for further pooling and so on. Similarly, available global production seeks the cheapest production sources taken in conjunction with available transportation efficiencies. Reversing the hydrology model to accomplish economic growth in developing nations is more difficult. Consumption of more than basic essential goods is being pushed uphill, dependent on rising wages and similar factors. That effort takes longer, as incomes are slower to rise than offset by downhill production shifts from developed to developing nations. Meanwhile, wages are headed downhill in the developed nations as production of goods and some services flow out of such nations. Basic hydrology.
There is no mystery as what is occurring. The public is treated to poor explanations by economists who are baffled by surplus capital, declining consumer savings in developed nations, and slow consumer demand in developing nations. Each of these self-imposed global trade constraints should have been anticipated by economists if they understood the basic WTO trade model.
A considerable portion of WTO trade model is basic hydrology. Nothing more. Economists would be well advised to grasp the concept, thereby improving their collective abilities to project outcomes in a more timely manner.
Most children in the world learned basic hydrology early on. It's a science that should not be overlooked in revised economic models and cognitive economic thoughts that are supposed to be capable of predicting economic outcomes based on the growing absence of tariffs.
Posted by: Movie Guy at September 28, 2005 07:51 PM
We need to forget about outsourcing?
There is a bloody elephant in the room--and we are not supposed to see it?
Do you actual watch what companies are doing or do you simply make things up as you go along?
I really suggest you do as movieguy suggests:
Look at the trade agreements. Actually look at the companies that offshored. Then look at the trade results.
For example, look at the growth of China's electronics exports. Do you really think those companies are Chinese? Where do you think the Ipod is made? Where do you think all the parts that go into a Dell are made or assembled?
Posted by: Stormy at September 28, 2005 08:29 PM
I'm sorry, but the king has no clothes.
We can all see that the US CA has huge holes in its underwear, namely:
*US wages cannot compete with the low Chinese wages,
*unstable and high oil prices, and
*a budget deficit, provoked by a war and others.
" Regarding Humid's query, that's the topic of a forthcoming Council of Foreign Relations Special Report ( http://www.cfr.org ) I'm writing. So I shouldn't give away the punch line. However, suffice it to say I believe the view that the U.S. CA deficit is primarily and exogenously foreign driven is wrong, and that U.S. fiscal policy *can* be used to shrink the CA deficit. "
I hate to punch your punch line, but the Chinese low wages and oil price crises, wich are both exogenous outweight the endogenous budget deficit. Or, are you thinking in terms of promoting a US dollar devaluation, which I find appropriate, under the circumstances.
If not, you seem to brush away the Chinese low wage issue --and Indian et al, as well--or the trade deficit relevance in the CA deficit. You brush away their ---- 34% yearly export growth for 2005!!!
My goodness, ...we cannot cover the sky with our hands.
What will it take for you to see that the projections indicate that in a few years we will live in a society similar to the Spaniards in the XVII century, rich but non productive; it lasted till their riches were gone...then what...services??
Let me give you some examples, the Chinese are already buying companies in the USA, keeping sales and minimal US manufacturing staff; this way, buyers think they're still purchasing US products, when in essence manufacturing is Chinese and they're dealing with a shell front --to service US customers. Isn't this what Walmart is all about, the largest company in the USA. And, yes, ...their employees provide the service.
You brush away the 1.3 billion Chinese population. It has been estimated that only 500 million participate in their new economy. Let me see...that leaves 800 million very unrestful sideliners; and I haven't even mentioned the 1 billion Indians, nor the 500 million Russians, nor the 500 million East Europeans.
That's quite a chunk of untapped resources, labor in this case, --and if we've learned anything from history-- FDI capital will flow to these virgin resource areas.
Under these circumstances, what kind of an idiot would invest in the US? Especially, if this investor thinks that the US dollar has to fall. You see, that's the sensible thing to happen in order to have low US competitive wages, which is the CA's only solution.
Low wages in the US is the only way to balance the CA because it fosters exports and hinders imports, and it further attracts investment --not the Mortgage Backed Securities kind, but the high quality job generating manufacturing investment.
The foreign CB US bond buying will continue, it is a good way for the Chinese et al to buy market share. And, as a full employment tool, it further makes sense for them to invest in US long term bonds.
Let's back up a little, untapped labor resources, which are vying for US markets --add to approximately 4 billion workers. This should put a downward pressure on labor prices worldwide; isn't this already happening?
If money is a proxy for accumulated labor hours, and we accept that wages are falling; then, interest rates must fall since interest rates represent the price of money.
So, interest rates should fall to account for the incorporation of this huge new labor resource.
This is why Mr. Greenspin will not be able to sustain his Fed funds rate increase.
Looked from a different perspective, a rising US dollar which is the consequence of rising rates hinders US exports and fosters imports, which exacerbate the CA deficit.
And a rising CA deficit should weaken the US dollar, which would make more expensive the cost of expanding market share for China et al; inflation should be felt in the world, and most harshly in China due to the low wage component in prices.
If Mr.Greenspin insists on inverting the curve; what about the carry trade --borrow long term, lend short term? This should rapidly flatten short term rates down to the long term level.
* Let the US dollar fall, the cost for the Chinese to peg their currency to the US dollar will be unbearable, and eventually US wages will become competitive as well as US exports,
* End the war, and
* Have the Treasury scare the shit out of oil speculators by shorting energy futures.
The rest is wishful thinking.
Posted by: Joe Rotger at September 28, 2005 10:52 PM
Menzie, what you said confirms what I was thinking: the US deficit and the Asian surpluses being higher than expected suggests that they are interconnected. On policy level this means that it is not easy to do something on this unilaterally.
And I really think that the underlying reason for this is the Chinese growth and very high investment and savings rate. And behind this is - and I think here in Econobrowser this view may be allowed - the Chinese energy production. It is not only or mainly low wages. Low wages are everywhere but only China is growing at this pace. A 94% energy self-sufficiency tells something especially if the domestic energy production is growing about 10% a yeart. The scale and pace of this is extraordinary, nothing like this has ever happened in the world. This cannot be a coincidence. If this is so it explains a lot - why it is "different this time", why the "deficits dont't matter" Goldilock economists have ridiculed the "pessimists".
China invests a lot in heavy industry and infrastructure. This is really very energy-intensive. How could the US increase its power supply 10% a year? In the moment Americans are happy to keep it at present level. So are the British.
The main Chinese energy source is coal. They have astonishingly been able to increase the production roughly 10% a year. But this means doubling it in 7 years. Now the coal production is at the level of 2 billion tons. It is easy to see that it will not be 4 billion tons in 2012. Will it be 3 billion tons? May be, but before that the growth rate has to slow down. To keep up the overall economic growth the Chinese must import more energy (oil, coal, natural gas). This will affect the trade balance. But also world energy prices.
It might well be that the expected slowing down of the Chinese economy has already started. This means that the investments and saving rates will also go down. Then we have the unwinding of the Asian-US imbalance system. The big question is, can the petrodollars save it. My guess is no, nothing will save the investments and world growth.
Posted by: TI at September 28, 2005 11:45 PM
Thanks to all for the interesting comments.
I confess: I am hopelessly mired in mainstream (New Keynesian) macroeconomic thought. That being said, I never claimed that everything was fine. In fact, my first post asserted that the U.S. current account deficit was driven by several factors that were unsustainable, and adjustment to smaller current account deficits could be painful.
That is why I suggested action to alter the course we are currently on. Yes, if we keep on spending profligately, if we fail to save (in both the private and public sectors), then over the longer term, our standard of living will suffer. Greater indebtedness will mean that our fate is to some extent placed in the hands of foreigners, or foreign governments. A greater stock of U.S. government debt leads to either a risk premium on dollar denominated assets, and/or great temptation by the Fed to inflate away our debt, thereby threatening the dollar's status as a key reserve currency.
At the same time, we shouldn't let anecdotes constitute analysis. China is important, but it still constitutes a small portion of world output, when evaluated at market exchange rates. The impact of China on prices is important but not primary. See Kamin, et al. http://www.federalreserve.gov/pubs/ifdp/2004/791/default.htm . In addition, rapid growth from China in exports is matched by quite rapid growth imports. China Economic Quarterly reports 2004 Export and Import (nominal) growth at 35.4% vs. 36% in 2004. 2005 should see much slower import growth. 2006 will depend upon what happens to the Rmb.
In addition, one shouldn't confuse absolute advantage with comparative advantage. Low wages give China comparative advantage in goods that use unskilled and medium skilled labor intensively, but they can't have comparative advantage in all goods.
Finally, I agree that the current account deficit cannot be solved unilaterally. It will require fiscal restraint at home, a serious program of energy conservation (either through taxes or by CAFE standards) and effective negotiations to allow greater East Asian currency flexibility. This last step will accelerate the adjustment of long term interest rates toward equilibrium which will in turn increase private savings in the U.S.
I lay out these arguments in greater detail at:
The summary of the Report is available at:
Posted by: menzie chinn at September 29, 2005 06:31 PM
A great discussion.
For clarification and understanding of my economic hydrology model as applied to the WTO trade model (its body of policies and conditions), I view China as a production clearinghouse. The best available labor rates, large and reasonably efficient factories, further capacity for more production, and so on. The lowest cost production pool.
Certainly, China is only the tip of the new trade regime in terms of where future production will flow if unimpeded. But we can rest assured that more U.S. outsourcing will occur until such time as trade policy or currency valuations are modified. No one wants to touch trade policy, so we're left with the hope that a U.S. dollar devaluation will address the some of the U.S. global trade imbalances.
I agree with your remarks about absolute versus comparative advantage. I look forward to noting what comparative advantages the U.S. will maintain as we drain our higher production cost pool into downstream offshore lower production cost pools under the cover of WTO trade policy.
Each time we lose an industry or a portion of such production, we not only lose the manufacturing GDP input but also the related supporting supply and logistics chains and their economic impacts. The overall losses are large.
The WTO trade model is worthy of a separate post. No main blog economist has touched a thorough explanation of WTO policies and member obligations in the past year to my knowledge. Few appear to be overly familiar with its provisions.
I certainly place its effects at the base of the origins of the U.S. current account deficit. If only others would be willing to discuss the WTO trade policies.
Perhaps one day...
Thanks for a fine discussion.
Creator, Economic Hydrology Theory
Posted by: Movie Guy at September 29, 2005 09:27 PM
Menzie -- impressive response to comments. Are you sure you do not want your own blog. Gotta check out your current account papers listing.
A few points:
1) To support menzie's mainstream new keynesian macro analysis of the trade balance, if you go back and look at the effects of NAFTA and the Uruguay Round, in broad terms, they triggered rapid growth in US imports and US exports but no expansion of the trade deficit. The big swing in the deficit started in 97, with the macro shocks in Asia, and then the new economy boom in USA. Menzie -- I am not as convinced as you are that the US competitive advantage lies in services. I agree with Dan Gross -- the big US service exporter, Hollywood, looks like a dinosaur, with a bloated cost structure and mediocre products. US exports of financial services (US workers intermediating between Chinese savings and Chinese investment) have never been as impressive as US exports of financial claims. We will see tho.
2) re: oil. Am working on this now. the difference in oil exports for the core oil exporters (OPEC plus Russia) for oil at 60 v. oil at 30 is around $440b. Some of the windfall has been spent, but most has been saved -- so my perception is that China and the oil exporters both are accumulating around $300 b in financial claims on the world this year. But compared to China, relatively little of that surplus shows up in formal reserves. Look at the reserve data on the Saudis web page.
3) Re: China. We now have trade data through August, and China's trade surplus is running at around $10b a month (sa) -- so this year's trade surplus will be around $120b. FDI inflows are around $60b, even with some outflows, China's basic balance will be close to $180b, or 10% of GDP -- even with high oil prices. Hot money flows explain a much smaller share of China's surplus this year than last. I also --as you know -- have a slightly different take on China from 02-04, when its trade surplus was comparatively modest. Specifically, given that China was experiencing an investment boom comparable to the US in the late 90s or the Asian tigers in the mid 90s, the real question is why China did not move into deficit. The absence of a deficit then set the stage for a large surplus now, when investment growth has slowed to a more normal rate (one interesting tid bit -- money growth has picked up strongly in China recently ... and there are some signs of a pick up in residential investment as well). China's overall exchange rate has not been stable -- it has moved with the dollar, so it depreciated in real terms in the face of a boom. Hence China's exploding surplus with Europe. Recent $ appreciation should slow that a bit ... all in all, my sense is that the break in dollar appreciation in 01 pushed china's XR substantially out of line. what worked from 95-01 (a rising dollar matched rising chinese productivity) stopped working, but the policy of pegging to the dollar did not change. the lagged impact of $ (and RMB) depreciation is still showing up (big time) in the 05 data -- see chinese export growth to europe, where domestic demand is kind of stalled.
4) Re the Fed's Sigma model. I have a few questions that you might be able to help me with (comparative advantage). As I understand it, the model basically postulates a degree of crowding out, which is why a $1 increase in the fiscal deficit only leads to a 20 cent deterioration in the current account deficit. The basic channel is that a fiscal expansion leads to an increase in real rates, a fall in investment and (i think to a lesser degree) a rise in savings. the model also seems to assume that part of the fiscal expansion is reversed over time, limiting its long-run impact. The increase in real rates leads to an increase in the real exchange rate, which formally leads to the increase in the trade deficit. But because trade responds weakly to real exchange rates, that effect is small. I hope I described the basic parameters correctly. My concern -- basically, none of the postulated channels worked as described in the 02-04 period. real interest rates did not rise. the real exchange rate did not rise -- the dollar fell v. the major currencies, stayed roughtly constant v. asia (whose currencies were quite depreciated as a result of the 97 crisis) and rose v. latam (brazil's 02 crisis). overall, the broad dollar fell. Shouldn't the absence of mapping between the model's channels and the observed results reduce our comfort in the model's results?
Bernanke argues that barring the fiscal deficit, real rates would be even lower, housing prices higher, private savings lower, etc -- but that is an outcome that is equally at odds with the model. His basic thesis seems to be that a shift toward savings in Emerging markets (or a fall in investment) generated excess savings that lowered real rates globally -- and the deficit worked to offset that impact. Are there any formal models that would incorporate that -- some global shift toward an excess of savings v. investment that interact in complex ways with a shift in US fiscal policy?
sorry about my long comment!
Posted by: brad setser at September 30, 2005 09:00 AM
WTO replaced GATT on January 1, 1995. The subsequent acceleration in trade barrier reductions resulted in the growing shift to larger trade and current account imbalances.
U.S. Goods BOP:
Posted by: Movie Guy at September 30, 2005 01:00 PM
the big swing did not start til 98, and i would argue it reflects:
a) a big fall in asian (and latin and russian) exchange rates v. the $
b) a collapse in investment in the asian nics
c) a rise in investment in the usa
d) soaring stock valuations that led americans to cut back on old fashioned savings ...
Posted by: brad at September 30, 2005 03:00 PM
Well, plenty of material to deal with here. Don't know that I can do justice to it all. But thanks to Brad for the details on what's happening with the oil dollars.
Let me focus on what I think I know something about: the nature of these dynamic stochastic general equilibrium models. They've all got at their core the new open economy macroeconomics structure laid out in Obstfeld and Rogoff's textbook. In order to match the data, or to be able to answer certain questions, frictions are introduced. The Sigma model includes a certain proportion of rule-of-thumb consumers that consume their income; hence as income changes, consumption changes along with imports. In the long run, everybody is Ricardian, so fiscal policy has no impact. The IMF's model introduces an explicit link between government debt and foreign debt motivated by a particular form of overlapping generations.
Do either of these models have the ability to predict what happened in 2002-04? Not really. But I'm not really sure that any formal model would be well suited to explaining the stock market boom and bust, and the extraordinary monetary policy pursued thereafter. If you believe there was a bubble in the stock market, and then in the real estate market, then this would not be surprising, since such models rule out bubbles.
Is there a way to explain the pattern of events that inspired Bernanke's "global savings glut" speech? I think part of it what Brad has mentioned -- the boom in household consumption due to stock market and real estate market bubbles. Part is the increase in corporate savings that increased the stock of world savings. Part of it is the fiscal deficit. These are flow explanations. On the stock side, there is the post 2002 accumulation of U.S. Treasuries by Japan and then China that further depressed long term interest rates, thereby maintaining the housing bubble.
On the role of services exports: I too used to think that services were too small a component to be important; but now that I think of vertical specialization, and the fact that there is double counting in goods imports and exports, I believe that services can be important in the adjustment process. But I admit, I'll have to do more analysis in order to be able to be definitive in this regard.
Posted by: menzie chinn at September 30, 2005 06:27 PM
three gap framwork budget deficit and current account deficit
Posted by: jacob gamu at May 10, 2006 09:48 AM