The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

« Energy Shocks And Monetary Policy: Can The Fed Ease The Pain? | Main | On That End To Germany's Soft Patch: Never Mind »

August 24, 2005

Global Dollar Demand And The U.S. Housing Market

This comes from today's Wall Street Journal (page A1 in the print version, subscription required for the online version):

Strong demand for mortgage-backed securities from investors world-wide is allowing American lenders to make more loans -- and riskier ones -- in a way that is helping prolong the boom in U.S. house prices.

The cash pouring in -- not only from U.S. investors but increasingly from Europe and Asia -- keeps stoking the housing market even as the Federal Reserve Board continues to raise interest rates, normally something that damps home prices. The market has shown a few signs of slowing recently, and talk of a bubble has grown louder, but prices continue to rise or remain at lofty levels as investors continue to gobble up mortgage-backed securities and banks keep lending.

This is but a variation on a theme I have emphasized in the past:  The "interest rate conundrum", the resulting stimulus to consumer demand, and the stubborn refusal of the U.S. current account to reverse course are all a piece of developments that are fundamentally driven by the desire of foreigners to send their financial capital to the United States.

A few comments about alternative opinions on the low-interest-rate/trade-deficit/consumption-binge troika.  Several of my fellow bloggers -- Brad DeLong, pgl, William Polley, and Mark Thoma -- commented on an article appearing a few weeks back in The Economist, sugggesting that an explanation that hinging on excess creation of liquidity. Several of these commentators were skeptical about The Economist's position.  Count me in on that. Although this will seem hopelessly old-fashioned, here is the recent record on money creation in the United States:


Broad money growth in the 4-6 percent range with nominal GDP growing at a 5-6 percent annual rate just doesn't spell excess liquidity to me.

In previous posts, James Hamilton has taken me to task for downplaying the trend in U.S. personal saving rates.  On this, let me just say that I think the differences between Professor Hamilton and me are not that large.  I agree that saving rates would be low even without the willingness of foreigners to finance American consumption.  My only point is that global influences have accentuated this tendency, and are impeding some of the adjustments that most of us think are inevitable.

Finally, pgl continues to assert that the total picture cannot be understood without gettting to the bottom of what he calls the "investment bust."  On this, my friend and I are on exactly the same page.

UPDATE: Tim Duy and I are on the same page as well.

ANOTHER UPDATE: The Big Picture covers this story too.

August 24, 2005 in Housing , Interest Rates | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Global Dollar Demand And The U.S. Housing Market :

» Excess Liquidity v. Credit Bubbles from Economic Dreams - Economic Nightmares
In a recent post at macroblog, David Altig makes the point, “Broad money growth in the 4-6 percent range with nominal GDP growing at a 5-6 percent annual rate just doesn't spell excess liquidity to me.” Considered in isloation, it [Read More]

Tracked on Aug 25, 2005 4:46:30 PM


What would happen if Freddie Mac and Fannie Mae never accumulated the 1.5 trillion portfolio of mortgages? They are funding these mortgages with short-term money, hedging their risk and avoided putting 1.5 trillion into the long end of the market. Just think where the yield curve would be if the bought back their short term funding and securitized their mortgages and sold them. My guess is that the yield curve would be steeper than it is today.

Posted by: Ed charlip | August 24, 2005 at 08:26 AM

RE: "Broad money growth in the 4-6 percent range with nominal GDP growing at a 5-6 percent annual rate just doesn't spell excess liquidity to me."
Isn't it reasonable to look at money creation for the last two or three years, instead of just one? And, shouldn't we question the leverage applied to the money? What if a lot of that 6% money-increase quickly found its way into hedge funds who leveraged it hugely, & money-lenders who eased home-lending practices to absurd levels? I'm concerned with the exponential growth of our derivatives market. How much is money growth spurring this unregulated leveraging & how much of this can our economy really support?

Posted by: bailey | August 24, 2005 at 08:48 AM

Borad money growth in the 4-6 percent rage is the current number. Where did the liquidity created between 1995-2003 go?

Posted by: Secret Admirer | August 24, 2005 at 11:53 AM


I am riding on a limited express, one of the crack trains
of the nation.
Hurtling across the prairie into blue haze and dark air
go fifteen all-steel coaches holding a thousand people.
(All the coaches shall be scrap and rust and all the men
and women laughing in the diners and sleepers shall
pass to ashes.)
I ask a man in the smoker where he is going and he
answers: "Omaha." Kindly visit
The Economic Fractalist http://www.economicfractalist.com/

Posted by: gary lammert | August 24, 2005 at 01:47 PM

While it is true that the weak aggregate demand growth from 2000Q3 to 2003Q2 was due to an investment bust, I agree with James Hamilton on the long-run issues raised by the low national savings rate.

Posted by: pgl | August 24, 2005 at 03:23 PM

* Broad money growth in the 4-6 percent range with nominal GDP growing at a 5-6 percent annual rate just doesn't spell excess liquidity to me.

Maybe I attend to this stuff too infrequently, or am too ignorant, but it seems to me that in today’s environment one needs to look at indicators of “excess liquidity” more broadly. Maybe you do as well, David, but it doesn’t jump out to me from this post. Here are snips from Doug Noland's latest "take," from http://prudentbear.com/archive_comm_article.asp?category=Credit+Bubble+Bulletin&content_idx=45778

“Broad money supply (M3) expanded $5.4 billion to a record $9.779 Trillion (week of August 8). Year-to-date, M3 has expanded at a 5.1% rate, with M3-less Money Funds expanding at a 6.4% pace. For the week, Currency declined $0.7 billion. Demand & Checkable Deposits declined $9.4 billion. Savings Deposits dipped $3.8 billion. Small Denominated Deposits rose $3.5 billion. Retail Money Fund deposits fell $1.9 billion, while Institutional Money Fund deposits added $0.5 billion. Large Denominated Deposits jumped $10.8 billion, with a y-t-d gain of $147.4 billion (22% annualized). For the week, Repurchase Agreements gained $6.4 billion, while Eurodollar deposits were unchanged.

“Bank Credit rose $5.5 billion last week. Year-to-date, Bank Credit has expanded $549.5 billion, or 13.2% annualized. Securities Credit declined $11.3 billion during the week, with a year-to-date gain of $133.9 billion (11.3% ann.). Loans & Leases have expanded at a 14.3% pace so far during 2005, with Commercial & Industrial (C&I) Loans up an annualized 17.8%. For the week, C&I loans declined $1.9 billion, while Real Estate loans expanded $7.8 billion. Real Estate loans have expanded at a 16.6% rate during the first 32 weeks of 2005 to $2.802 Trillion. Real Estate loans were up $380 billion, or 15.7%, over the past 52 weeks. For the week, Consumer loans added $1.7 billion, and Securities loans jumped $10.5 billion. Other loans dipped $1.2 billion.

“Total Commercial Paper jumped $13.3 billion last week to $1.588 Trillion. Total CP has expanded $174 billion y-t-d, a rate of 19.4% (up 17.7% over the past 52 weeks). Financial CP surged $13.7 billion last week to $1.446 Trillion, with a y-t-d gain of $161.8 billion (19.9% ann.). Non-financial CP dipped $0.4 billion to $141.8 billion (up 15.0% ann. y-t-d and 9.9% over 52 wks)."

Here is a link to Noland’s encapsuled view of the emergent world, titled “Contemplating the Evoluation from the Way We Were to the Way it Is?” http://www.prudentbear.com/Bear%20Case%20Library/bear_case_library_images/Contemplating_the_Evolution.pdf

One snippet:

“Nowadays, Federal Reserve operations work mainly by aggressively manipulating rates, yield spreads and, increasingly, market perceptions. In the process, the new 'tool kit' bolsters leveraged speculation and unparalleled Credit Availability. The key analytical point is that the effects of today’s operations are so much more disparate and unpredictable than when the Fed was managing the bank reserve 'anchor.' I would argue that the Fed is now held hostage to the markets – financial and real estate – and particularly to endemic Credit market leveraged speculation. In short, financial and central bank evolution has parented a mutant, uncontrollable Credit System. There is no financial system anchor, while the Fed’s new 'tool kit' emboldens speculators and nurtures excess.”

Can one be comfortable these days only looking at M3 when talking about “liquidity?” Aren't these other credit expansion indicators worth considering as well? Noland and others argue that the creation of liquidity is much broader than what central bankers monitor. And that, he contends, is a big problem for us all.

Does Noland's thesis have any merit? If not, why not? I’ve been looking to debunk Noland’s thesis for several years, but haven’t yet found a credible counterargument. Can any here provide one?

Posted by: Dave Iverson | August 24, 2005 at 05:18 PM

I'm not an economist, but I have two comments re: money supply.

The first is whether M2/M3 are valid. I suspect that they are. Even if they are a smaller proportion of money than in the past, they may still be the marginal supply of money. In other words, shrinking supply deprives someone of credit, which reduces someone else's income, etc. To say it is a smaller engine does not mean it is not the head locomotive on the train.

The second pertains to Noland's comment on M2/M3 being obsolete, and that the GSEs (Fannie, Freddie, etc) are creating money. It is true to my understanding that when the GSEs issue paper, this paper is valid collateral as far as the Fed is concerned. It is equally good as Treasury debt inasmuch as the Fed will loan you reserves, ie dollars, based on either security. So mortgage lending inflates bank assets.

But GSE paper is not legal tender. To actually buy anything requires dollars. This requires a financial institution to borrow from the Fed against GSE collateral. This is indeed captured in M2/M3, no? Thus it appears to me that financial system assets may be overstated, but not money supply.

Even if there is no direct link between GSE paper and money supply, there is probably some link somewhere. People who have credit, or believe they can get it, will overspend their means. Looking for evidence of causation may be interesting to an economist, but it doesn't seem necessary to make an informed decision. And that is that people are overspending which will have to correct at some point.

Posted by: chris | August 25, 2005 at 12:51 AM

Finally, pgl continues to assert that the total picture cannot be understood without gettting to the bottom of what he calls the "investment bust." On this, my friend and I are on exactly the same page.

Review Schumpeter's work...

The reason money is flowing here from overseas is they feel there is little 'worthy' to invest in over there... Once it gets here we feel there is little 'worthy' to invest in here... except possibly real estate either directly or indirectly (via MBS)... It truly IS a global investment bust.

That is completely consistent with the tail end of innovation driven long business cycles described by Schumpeter in the '40s.

I am not saying Schumpeter has 'all the answers' but I can tell you the innovation driven long wave stuff explains a lot of what went on in the 90s and is going on now...

We have come off of, or are just past peak, one of the worlds truly great innovation surges... from say early 1980s to say late 1990s... basically the convergence of silicon computer processing & telecommunications. Result: the internet and with it real-time globalization.

It is as big as the iron-coal-steam power innovation wave of the middle-to-late 1800s (think railroads, steamships & the Guilded Age) or the internal combustion engine & petroleum innovation wave of the early-to-middle 1900s (think automobile, airplanes & the Roaring 20s).

The internet-globalization revolution we have lived through is every bit as big a deal.

Everyone of these previous 'waves' generated monstrous amounts of wealth... real wealth & real productivity gains for real people... wages & investment returns & standard of living gains.

Unfortunately they also all ended in 'busts' - the Panics of the late 1800s & the Crash of 1929. The two unfortunately are related.

I think we are well on the way to another bust - near miss with the NASDAQ - unless we can figure out a way to moderate the current trends.

The problem is that all that wealth created during the up slope of the innovation wave is looking for a home. But now we are on the down slide in innovation activity. What just recently were innovative products are converging - becoming standardized commodities. The primary battlefield for commodity products is 'cost & price' not 'features & performance' like during the high innovation part of the cycle.

To have an advantage in a commodity market, companies get bigger & more vertical... mergers & layoffs result. The largest, lowest cost survivors do well - at least for a while - but many others are squeezed out. As pressure mounts - the producers search for even lower cost production options - lower capital & labor cost areas - so as to compete with their nearly identical competitors & their offerings. It isn't difficult to relocate these products because by this time in the cycle they are well understood and nearly identical worldwide... eventually near the very end, even the largest lowest cost survivors are struggling with low margin, low profit... the race to the bottom is near the bottom then.

The first losers in this process are the well paid workers who came up during the early part of the wave... commoditization squeezes them out pretty early on (we see that with western IT workers now).

The next major problem is for investors. Where do you invest new money or even previous gains in such a climate? Buy into the commodity industries? Most of their growth is behind them. Equity prices are high and relative return lousy. New start ups? Maybe if you can find them but it is too late for 'new' start ups in the established industries - entry is far too painful & difficult... And the 'new' new start ups outside the established industries aren't well understood... many really aren't ready for prime time yet... don't have either a product or a market to sell into yet.

So this pool of investment money hunts for rates of return it is accustomed to and instead of finding productive activities... finds speculative ventures & scams. Shakey financial instruments & speculative real estate that eventually ends up in a bust... we are heading there now.

So what do we do about it? I'm not sure at all. Three things come to my mind:

1. Change policy to try and hurry up the next wave of innovation. There needs to be a big increase in levels of basic R&D (no one knows where the next big wave will come - so build a solid base of understanding and let the synergies & serendipidies happen).

2. Change tax law & financial regs to favor REAL investment as opposed to short term speculation. I think aggressive depreciation reform is the key... even for intangibles like goodwill... But retain a fairly high cap gains rate to keep the incentive on investment in & not cashing it out. If the depreciation schedules are favorable enough... there will be PLENTY of incentive to overcome future cap gains tax bills...

3. Safety nets. Tail end of these innovation cycles are very disruptive & painful (the creative destruction phase). I don't see how they can ever be completely avoided. However when trouble hits, if people are fed & housed instead of driven into Hoovervilles... then the socio-political effects are less and policy makers aren't having to fight riots, revolutions & world wars while simultaneously trying to get the economy taking off again.

In summary - you want to get a handle on the 'investment bust'... review Schumpeter's work and modernize it for todays conditions (not mid-20th century). JMHO.

Posted by: dryfly | August 25, 2005 at 11:38 AM

The housing bubble is the finale of a seventy year credit cycle of immense proportions:

Saturation Curve Fractal Analysis - A Real Science?

In order to qualify as a true science, the subject entity must betestable by scientific method and have underlying laws that operate in the
real physical environment. These laws must be repetitively provable and have
reasonable predictability for different applications. Scientific testing in
college biology, chemistry, and physics laboratories usually results in
experimental values that roughly support the underlying mathematical
equations and theoretical constructs. If indeed complex economic systems
travel by the simple quantum laws that observational fractal analysis
suggests, a similar validity should be testable and provable in the great
laboratory of readily obtainable asset valuation saturation curves.

Valuation fractals represent a composite integration of primarily six
elements in the complex economic system: cash and savings; total private,
corporation and governmental debt load; ongoing wages; assets; lending
practices; and prevailing interest rates. Each of these six broad parameters
has its own complex internal dynamics and summation characteristics. In a
very mechanistic fashion, following simple near-quantum and near-quantum
related Fibonacci numbers, valuation fractals 'grow' to buying saturation
levels and thereafter 'decay' to lower selling saturation levels. The
fundamental point to this new potential economic science is that the daily,
weekly, monthly, and yearly valuation fractals represent the sum total
integration of those six elements and their complex interactive
relationships. Pour into the economic vat: cash for daily transactions,
savings available for money to be borrowed at given interest rates using
prevailing lending practices for both major purchases and minor credit card
purchases, balanced by on-going wages and debt servicing obligations,
balanced by relative valuation of assets and their relative state of
consumption, mix it up on a daily, weekly, etc. basis - and - from the vat
flows forth the daily, weekly, etc. summation saturation curves dancing to a
rather precise near quantum fractal tune. While lower order time unit
fractals such as minutes and hours represent trading valuation saturation
points, intermediate fractals represent the larger picture of on going
velocity of money growth percolating through the system. The higher order or
4-yearly, 17-18 yearly and 70 year fractals represent both business cycle
and asset and debt saturation levels at the basic consumer level.

There are three sequential identified ideal growth fractals followed by a
decay fractal. The near quantum number time units for the three cycles are
x, 2-2.5x and 2x, respectively. A nonlinear devaluation typically
characterizes the second growth fractal somewhere between the 2x and 2.5x
time period. The third growth fractal which ideally is 2x in length can have
an extension to 2.5x. This extension of the third growth fractal has
characterized both the current US equity and heavily invested commodity
areas, particularly oil and gold, for the entire 128 week duration of the
March 2000 secondary growth period.

Just as the complex system is an integrative process, valuation fractals
which exactly represent them are likewise composite integrations with
nonlinear capacitor like decay devaluations. Fractals incorporate the
terminal portion of the preceding decay fractal into the beginning of the
follow-on growth fractal. An elegant pristine example of this rolling
integration was the 40/100/100 day cycle exactly x/2.5x/2.5x that resulted
in the March 2005 top for the DJIA. The first two fractals were 'declining'
growth fractals with a very characteristic nonlinear break at the end of the
second fractal in August 2004. That second fractal was likewise elegant in
its evolution in that it was composed of a 29/72 day x/2.5x sub fractal
sequence. The probability that these precise sequences are random numerical
sequential events approaches zero and elevates fractal analysis,
reciprocally, to a high probability real science descriptive of the complex
macro economy.

The subsequent growth fractals dating from August 2004 likewise have
followed the same very precise fractal growth evolution with a 52/130
(x/2.5x) day first and second fractal growth sequence with the typical
nonlinear drop between 2x and 2.5x of the second fractal. Anyone can verify
this pattern using any of the major US or European indices. The third
fractal US equity sequence has been a 12/30-31/28 day sequence, approaching
the extended ideal form of x/2.5x/2.5x growth pattern. The major European
indices ,e.g., the FTSE, DAX, and CAC have a slightly different mix of the
six aforementioned underlying elements and have extended their growth - but
are still confined within the 52/130/104 theoretical maximum and the
theoretical Fibonacci maximum of 52/130/(1.62 X 52 = 84-85)
days. These recurrent numerically ideal patterns since August 2004 once
again lend substantial credibility to the notion that the complex
macroeconomy operates according to some relatively precise laws of fractal

What are the rate limiting factors that result in growth saturation points
or asymptotes, decay selling saturation points or asymptotes, and the
general nature of fractal patterning? Each of the six controlling
parameters- assets, ongoing wages, lending practices, prevailing interest
rates, debt load, and cash and savings - contribute to the saturation areas.
Some are more important than others in determining cycle lengths and
saturation points.

Assets have two important elements: relative valuations and saturation
ownership. If the valuation becomes too high or too overly consumed, demand
will decease. The timing for this decrease is exactly represented by an
asymptotic valuation saturation level or a single high valuation point
followed by lower valuations. The valuation curves provide precise
'barometric' information on instantaneous demand relative to valuation level
and relative to the consumption level. Some assets such as gas and oil must
be purchased to maintain livelihood. As global consumption for the this
finite resource increases, resulting price increases squeeze the null saving
US consumer, far too many living from paycheck to paycheck, to the
financial breakpoint. Unnecessarily expensive US healthcare, 25 percent of
the value of which goes to third party insurers and the non-value added bill
collection system, can be considered yet another consumable asset, that,
like 'uninsured equivalent' gasoline prices, is driving many to insolvency.

Ongoing wages and just as important the jobs that support those wages are
perhaps the most important rate limiting factor in determining valuation
saturation points. In the US jobs sphere, high paying manufacturing jobs
with the exception of the housing industry have been significantly
outsourced. As the housing bubble crests, overcapacity will become evident
and high paying home construction jobs will contract. A considerable subset
of jobs in America have questionable value-added real economic worth and
will be lightened during consumer retrenchment. It is easy to image using
the 1930's as a template of a positive feedback contracting system, whereby
decreased ,e.g., construction jobs lead to decreased consumer spending which
leading to further job contraction in other nonessential service areas which
leads to further spending contraction and so forth.

Lending practices and prevailing interesting rates, the latter a Federal
Reserve controlled parameter, work in synergy to foster money creation and
asset inflation. Fractional reserve lending practices amplify the bank and
money market savings used as a reserve base for lending. Extremely low
interest rates, i.e., a Fed fund rate of 1 percent coupled with a lending
practice of LIBOR type loans, no money down and interest only payments
creates the interesting situation in which the interest cost of money is far
below the real asset inflation rate. Not to borrow is to lose money that
would be made with the expected inflation. Conversely, saving money under
these interest rate and lending practice guidelines results in loss of
purchasing power. Credit card interest rates reflect the needed higher
interest rates to overcome the default rate. The last year of higher Fed
Fund interest rates have resulted in both increased mortgage payments and
decreased bank profitability secondary to the contracting spread of long
term verses short term interest rates.

Ongoing debt load and the requirement to service that debt diminishes cash
available for asset consumption and investment. Percentage wise the total
debt load relative to wages and GDP has had relatively small incremental
increases - a fact which has mistakenly reassured many linear thinking
economists. Debt load becomes very important and a primary factor in the
fractal decay process, where assets are liquated in an attempt to pay down
debt. Because debt is in a major way based on the value of the asset, debt
load becomes relatively greater with ongoing declining asset values. This
process also represent a positive feedback system and is self perpetuating.
It results in a mechanistic devaluation and deflationary process, lowering
the value of nearly all non cash or non-cash equivalent assets.

Cash is the money that is represented by greenbacks in circulation and
greenback equivalent readily convertible debt instruments such as
treasuries, notes, bonds, bank deposits, and money market funds. In short
cash represents the dollars in circulation and savings. The savings rate,
which the Federal Reserve has bemoaned to be dangerously low and was
reported to be zero in July, reflects the competition of the the various
Investment areas. With interest rates below the real(which includes
housing) asset inflation rates, deposited money in saving instruments loses
its purchasing power value each week that it is malinvested in the bank or
interest bearing cash equivalent instruments. Deposited money in saving
instruments has been generally a bad investment in the last few years.
During the decay fractal process, this scenario will be reversed with money
from ongoing asset liquidation flowing into cash and cash equivalents, whose
purchase power value will increase relation to asset devaluation.

These are the lumped six broad elements that are dynamically interacting
with each other to create the summation valuation points, curves, and
saturation asymptotes. The evolving integrative fractals that appear to so
well describe the real instantaneous state, the trending state, the
saturation areas, and importantly predict the expected fractal
nonlinearities of the complex macro economic system, have the fundamental
characteristics of a real science.

Gary Lammert

Posted by: gary lammert | August 26, 2005 at 05:09 AM

Wow -- I take a day off and miss a lot. Let me try to address some of these (pretty terrific comments) as best I can in a brief space:
Ed -- The influence of Fannie and Freddie almost certainly influence the character of the housing boom. But although they facilitate things by making the market so liquid, I'm not sure we want to make them the cause.

bailey and Secret Admirer: Im not sure how long and variable those lags should be from money creationto their effect, but the graphs I presented really pick up the trend back into 2002 (because they are 12-mos moving rates). That's long enough for me. The data from at least 1999 through 2001 is distorted by Y2K. As for the mid-1990s prior to what we might reasonably attribute to Y2K, that's a fight for another day.

pgl -- agreed. I've been meaning to post on this for some time now,. Hopefully, I willget to it next week.

Dave -- It is clear that we are not that comfortable speaking of monetary aggregates as the object of our policy, since we don't really do it anymore. Still, I'm just not convinced it isn't the right way to think about Fed money liquidity creation, even if its usefulness as a forecasting tool per se has vanished. Here I would just piggyback on chris' comment above. There is no doubt that there is just a ton of liquidity sloshing around out there. But why does it make sense to relegate its existence to some new channel of influence that has been bestowed on the Fed from higher powers? Especially when you have plenty of very straightforward candidates in the form of rapid private financial-market innovation -- jut say derivatives and you are almost done -- coupled with the (perhaps strange) unwillingness of firms to fully abosrb the funds available?

dryfly -- Those dynamics may very well be the true ones (althoug I suspect I might view them a little less pessimistically than you.

gary -- well, you pretty much exhausted me. But I gather you feel you have a powerful forecasting model on your hands. What's it track record?

Posted by: Dave Altig | August 26, 2005 at 08:39 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad