Saving Glut Redux

Bernanke recaps his interpretation of the explanation for global imbalances. Is it any more convincing than the first time?


From his speech in Berlin on Monday:

Global Imbalances: Recent Developments and Prospects



In a speech given in March 2005 (Bernanke, 2005), I discussed a number of important and interrelated developments in the global economy, including the substantial expansion of the current account deficit in the United States, the equally impressive rise in the current account surpluses of many emerging-market economies, and a worldwide decline in long-term real interest rates. I argued that these developments could be explained, in part, by the emergence of a global saving glut, driven by the transformation of many emerging-market economies–notably, rapidly growing East Asian economies and oil-producing countries–from net borrowers to large net lenders on international capital markets. Today I will review those developments and provide an update. I will also consider policy implications and prospects for the future.


A principal theme of my earlier remarks was that a satisfying explanation of the developments in the U.S. current account cannot focus on developments within the United States alone. Rather, understanding these developments and evaluating potential policy responses require a global perspective. I will continue to take that perspective in my remarks today and will emphasize in particular how changes in desired saving and investment in any given region, through their effects on global capital flows, may affect saving, investment, and the external balances of other countries around the world.

There’s been some commentary elsewhere, by Brad Setser, Mish, Greg Mankiw, among others. For me, I will focus in on what I think are the most problematic aspects of his new discussion, and refer readers to other issues I’ve covered in the past, here: [1], [2], [3], and [4].


Consider this paragraph:


In fact, there is no obvious reason why the desired saving rate in the United States should have fallen precipitously over the 1996-2004 period. Indeed, the federal budget deficit, an oft-cited source of the decline in U.S. saving, was actually in surplus during the 1998-2001 period even as the current account deficit was widening. Moreover, a downward shift in the U.S. desired saving rate, all else being equal, should have led to greater pressure on economic resources and thus to increases, not decreases, in real interest rates. As I will discuss later, from a normative viewpoint, we have good reasons to believe that the U.S. saving rate should be higher than it is. Nonetheless, domestic factors alone do not seem to account for the large deterioration in the U.S. external balance.

First, as I’ve noted previously, arguments like this related to the budget balance should be banished from intelligent conversation. Or at the least, anybody who’s ever run a regression with more than two variables should not be saying pointing to two variables’ movements as a proof. This is why I don’t use the fact that “…between 2000 and 2005, there was approximately a 4.3 percentage point swing in the Federal budget balance, and a 2.2 percentage point swing in the current account balance” as proof conclusive of my twin-deficits perspective, but rather prefer to rely on regression estimates (see this post).


Second, I think that it is a debatable point that there is no obvious reason why saving rates should have decreased in the U.S. over the 1996-2004 period. In fact, given that real interest rates have been quite low in the early 2000′s, it’s a rather odd statement.


While I’m not going to assert that monetary policy was the cause of unnaturally low real interest rates (that in turn might have contributed to the housing boom and the associated mortgage equity withdrawal…), I will claim there is a plausible argument that the extended period of monetary ease was a contributing factor. It’s even more plausible, when one considers the rise in real (risk free) real interest rates even as China and oil exporting countries continue to run large surpluses after the increase in the target Fed Funds rate. This latter observation is documented in this post from January. (I’ll also observe the notion of real interest rates being equalized across borders with free capital mobility is not necessarily validated by the data — so low US real interest rates might or might not be indicative of anything, [5].) It all sort of depends whether you think relative PPP holds instantaneously; if you don’t think so, then monetary policy can cause divergences in real interest rates.


While my remarks might seem critical, I want to stress that I think there are some aspects of the global saving glut explanation (I prefer investment drought explanation) that are important; East Asian current account balances appear larger than what is predicted by historical relationships involving demographics and budget balances. And what is more important, in this speech Bernanke allows that other factors might be important in pushing interest rates low:


Once again, however, I do not want to rely exclusively on this line of explanation for the behavior of long-term real interest rates, as other factors have no doubt been relevant. In particular, term premiums appear recently to have risen from what may have been unsustainably low levels, in part because of the greater recent volatility in financial markets and investors’ demands for increased compensation for risk-taking.

To sum up, the Bernanke explanation for the US current account deficit relies upon a particularly small effect of budget deficits on current account deficits, and treats the US housing boom and associated mortage equity withdrawal as largely exogenous, or primarily a function of foreign excess saving. If you believe these points, then the saving glut story is the story for you.


By the way, Bernanke’s thesis still ignores — as Brad Setser cogently points out — the role of the official sector…

Technorati Tags: ,
,
,
housing,
twin deficits,
recession.

StumbleUponLinkedInReddit

19 thoughts on “Saving Glut Redux

  1. guest01

    dear Prof. Chin,
    This is rather very basic question, but should we look at net or gross capital inflows to the US to see *who* is financing US external deficit? Looking at net, it is clearly central bank’s reserve accmulation. Looking at gross, it is private sector. My intuition is to look at net obviously, but the IMF seems to be looking at gross figures, and so does Bernanke (not sure though)…

  2. bsetser

    Menzie — thanks for the plug.
    I am curious what you think of Dr. Bernanke’s argument that the “savings glut” v “investment drought” balance shifted more towards a “savings glut” in the 05-06 period, as the large increases in the emerging world’s surplus came from places where investment was rising (China, oil exporting economies) just not as fast as savings. I tend to find that argument relatively persuasive (though I would emphasize that it is a glut in “official savings”). But I get a sense that you don’t, and want to better understand the source of our different interpretations.

  3. DickF

    Menzie wrote:
    Second, I think that it is a debatable point that there is no obvious reason why saving rates should have decreased in the U.S. over the 1996-2004 period. In fact, given that real interest rates have been quite low in the early 2000′s, it’s a rather odd statement.
    While I’m not going to assert that monetary policy was the cause of unnaturally low real interest rates (that in turn might have contributed to the housing boom and the associated mortgage equity withdrawal…), I will claim there is a plausible argument that the extended period of monetary ease was a contributing factor.
    Menzie,
    You better sit down. I agree with you, though for a slightly different reason at least we agree.
    “…I have been able to compare currency inflation indicated by gold to the change in savings rate as indicated in the paper that the FED released of BB’s speech in Germany. What I have found is that BB and most economists miss the most important element to a national savings rate.
    “…savings have an inverse relationship to inflation, as inflation increases savings declines. This is not revolutionary but I have never seen it discussed by mainstream economists. …
    “I do not believe that the citizens of the country are actually changing their propensity for saving, but that they change their method of saving in an inflationary environment….
    “Economic saving is defined as disposable income minus personal consumption expenditure. That being the case the purchase of hard asstes are not included in savings…. For the average person the best way to maintain the value of wealth in an inflationary environment is to purchase hard assets that will appreciate. The asset that makes the most sense is real estate, especially a home.
    “In an inflationary environment those who own homes will purchase more expensive homes assuming that the inflation in wages will make the payments on the home easier and at the same time they leverage their “savings” based on the homes inflating asset price.
    “What my analysis showed is that those countries with the most stable currency have the greatest savings because there is less loss of value in the currency than when there is inflation.
    “…Almost everyone I knew in the 1970s used the same logic and real estate boomed…real estate booms are less dependent on interest rates than they are on inflation….the desire to protect asset value from inflation is a greater motivation than interest rates, even double digit interest on savings, as we saw in the 1970s. People make the calculation that any gain in interest will be eaten up by inflation.”

  4. Joseph

    Like many things in economics, I understand the mathematics of the current account deficit much better than the real world implications. Does it mean that the people of China are currently working their fingers to the bone to provide a better standard of living for Americans? Why would they do that? Is it because they believe that future generations of Americans will be working their fingers to the bone for the Chinese when the bill comes due?

  5. Phyron

    First, as I’ve noted previously, arguments like this related to the budget balance should be banished from intelligent conversation. Or at the least, anybody who’s ever run a regression with more than two variables should not be saying pointing to two variables’ movements as a proof.
    Seriously… you or Ben Bernanke in a room… who do you think is going to eat whom??

  6. PrefBlog

    September 13, 2007

    Goldman’s Global Alpha fund lost 22% in August:
    The fund, managed by Mark Carhart and Raymond Iwanowski, has dropped by a third in 2007 and 44 percent from its peak in March 2006. Investors notified New York-based Goldman last month that they pla…

  7. HZ

    I think the “investment drought” perspective is indeed quite a good one. If China, for example, had the right know-how, infrastructure and other conditions, it could have grown far faster domestically given its savings rate. It is playing catch up after all and does not have to invent things like railroads, electric power, telephone, airplanes, autos etc to drive the economy forward.

  8. knzn

    “unnaturally low real interest rates….the extended period of monetary ease was a contributing factor”
    This is just a Keynesian version of the savings glut hypothesis. Why was monetary policy so easy for so long? Because the global deficiency in aggregate demand (in other words, the excess of savings over investment) necessitated a monetary stimulus.
    In a full-employment world, one with “perfect” monetary policy (in the sense that monetary policy is always just sufficient to generate enough demand to reach but not exceed potential output), interest rates would perhaps have been even lower. Arguably, recognizing the role of monetary policy makes the case for the savings glut hypothesis even stronger.

  9. Laura

    Abstracting from any questions about the accuracy of the import price index, I was wondering if higher is bullish or bearish? We can ignore petroleum if that helps. Take the exchange rate as exogenous. Word on the Street is that you’re the expert on this sort of thing. Thanks.

  10. knzn

    Joseph, I think the relevant concern for the Chinese is not how hard they work but how many of them have productive jobs. If a Chinese peasant gets a job in the industrial sector, the improvement in living standard is presumably sufficient that the former peasant can give up a large chunk of the new income to provide a subsidy to US consumers and still come out ahead. In that sense, the low US savings rate — whatever its cause may be — has been largely a good thing from the point of view of the world economy, though it’s possible we have gotten to the point where the world economy is expanding too quickly, and the reversal of the low US savings rate may also be a good thing.

  11. Menzie Chinn

    bsetser: I guess I don’t have a problem with the conjecture that the rest-of-world saving increased more than rest-of-world investment in ’05-’06. My main problem remains the argument that the rest-of-world excess of ex ante saving over investment drove US real interest rates and drove the US current account balance, almost to the exclusion of other forces.

    guest01: I’d say in a world of perfectly fungible financial and physical capital, one should look at net flows. To the extent that some assets are imperfect substitutes for others, one might look to the gross holdings. But since financial assets are pretty fungible, and capital flows are dominated by financial flows, I’d say look at the net.

    Joseph: I don’t have an answer for you. In a standard intertemporal optimizing model with no information asymmetries, no commitment problems, and no government, it would be hard to explain why China is running a surplus with the rest of the world, given its low per capita income and likely trajectory in per capita income.

    One explanation forwarded is the Dooley et al. hypothesis that China wishes to intermediate its saving through the US, rather than through its own domestic financial system. There are enough criticisms of this view that I need not spend time here.

    Another view is that the Chinese government has decided that net exports is better “quality” aggregate demand than domestic investment, and domestic consumption. Whether this assessment is reasonable determines in part whether this is optimal for China to run a surplus.

    laura: Since I don’t usually think in terms of the implications for the stock market, it’s hard for me to say whether the import price numbers are bullish or bearish. I can say that the fact that the non-oil import prices didn’t move up means that this is less evidence of inflationary pressures — both because exchange rate pass through seems to be muted and because import prices rise (on their own) along with other goods prices, due to domestic factors as well. To the extent that this increases the likelihood of a Fed interest rate cut, it’s bullish. However, to the extent that pass through and inflationary pressures are muted, this probably buttresses the case for a slowdown in the economy, which in turn means lower earnings into the future — that I presume is bearish.

    knzn: I disagree. The main motivation for the long, prolonged, decrease in the Fed’s policy rate is to be found in this working paper.

    Phyron: I’m not sure I understand your point. Could you please be more explicit?

  12. Lord

    I really see savings glut and investment drought as two sides the same coin. Had there been valuable opportunities to invest that savings, it wouldn’t have gone into housing. The savings situation in China is partly due to its high aging rate and political situation is obviously a handicap to long term investment there.

  13. knzn

    I don’t see how the motivation suggested in the working paper about deflation is at all inconsistent with my explanation. Disinflation — and the possibility of deflation when disinflation goes too far — is caused by a deficiency in aggregate demand. Saying that aggregate demand is deficient is equivalent to saying that savings exceeds investment. If it hadn’t been for the savings glut, deflation would never have been an issue, and it would not have been necessary to pursue such an easy money policy.
    The only way this might not be true is if there was some exogenous shock to the inflation rate, so that it fell further than would have been implied by the level of aggregate demand. The one candidate shock I can think of is the increased availability of cheap goods from China, but I would see that as part and parcel of the savings glut. In that case it is public savings rather than private savings, but it is still savings: sovereign investment occasioned by exchange market intervention is still part of national savings.

  14. knzn

    There is one other possibility: that the Fed misinterpreted the inflation data by taking data on stagnant house rents as evidence of very low inflation rather than attributing them to the already low interest rates (which were reducing the annuity value of housing). In that case, there was a positive feedback mechanism that pushed interest rates further down than the savings glut alone would have. But it seems like a stretch to argue that this was more than incidental. Surely the Fed is smart enough not to get caught in that kind of feedback loop.

  15. David Beckwortgh

    knzn:
    Your arguement that the easy monetary policy of the past few years was an endongenous respone to weak world demand may be so, but why is it so hard to accept the possibilty that just maybe the Fed did something exogenously? I think Menzie is right. The Fed was a contributing player to the global imblances and the excess savings was in part a response to the loose monetary policy, not the cause (entirely).
    In short, I see room both for a saving glut interpretation and an (exogenous) liquidity glut interpretation of global imbalances. (I wrote a few posts about it here
    and here)

  16. David Beckworth

    knzn: you assume the deflation over this time was from weak aggregate demand. What about aggregate supply driven deflation? For example, I interpret the deflation scare of 2003 as a misunderstood case of productivity-driven deflation. Using a Wicksellian perspective, easing monetary policy when productivity growth is accelarting is distortionary. (I have a post on this too here
    )

  17. Buzzcut

    Another view is that the Chinese government has decided that net exports is better “quality” aggregate demand than domestic investment, and domestic consumption. Whether this assessment is reasonable determines in part whether this is optimal for China to run a surplus.
    It seems to me that you have two different models for rapidly industrializing an economy.
    You have the American model, which does so by driving domestic consumption.
    And you have the export oriented Japanese/ South Korean model, which does so by totally ignoring domestic consumption and relying entirely on exports for growth.
    The Chinese are following the Japanese/ South Korean model. Why are they doing this? Well, it certainly has worked for the Japanese and South Koreans! Maybe they think that following an Asian model is more likely of success than following the American model? Maybe they think that the American model occured too long ago, things have changed too much, and the more recent success of Japan and South Korea is again more likely of success?

  18. guest01

    Dear Prof. Chin, Thank you very much for your reply on net vs gross. Bernanke seems to be looking at gross to make an argument for global saving glut. But when we look at net, the large sourge of capital inflow to the US is official reserve increase….

Comments are closed.