Urban Survival - George Ure's weekly market update -- NEWS ONLY

Replaying 1929!

From February 10, 2001

Implications of The Mazurok-Ure Correlation: Multi-dimensional Economics

We're on the Titanic and that's a DEBTBERG©

The old saying is that a picture can tell a thousand words, so this week, I'm pleased to present for your viewing and trading enjoyment, a further set of notes, and some of the heavier math behind this thing called the Mazurok-Ure Correlation (MUC). Essentially Ehor Mazurok & I argue that changes in interest rates do not always have the same effect on the economy. This is due to how the underlying mass of debt and the economy is structured at a particular moment. This determines and where we are in the economic long wave. Sometimes, things work backward.

Last week, I told you how the economy behaves something like a wing of an airplane. A few readers had a tough time getting the concept, so I sat down at the big workstation on the boat and scanned in some quickie "cocktail napkin" sketches to help fill in the blanks. This first sketch below shows the MUC view of the economy as a multi-dimensional semi-solid. I think this has huge implications for the general field of economics. What we've done is used multiple discrete known dimensions to construct this solid that we call the "Debtberg". Reader Colin Seymour (who has an excellent news and econ site you should check out) advised us: " I hate to be the bearer of bad news, but according to this URL http://www.the-church-of-god.org/nw9802.htm the term "debtberg" was used by Ellen Spragins in the 2/23/98 edition of Newsweek". Still, we'll call 'em debtbergs and every economy creates one.

As you look at the Debtberg, you'll see a lower line that has years to give the Debtberg its time scale. The bottom of the berg is "pulled out" to reflect the average duration of mortgages. About the middle of the Debtberg you'll see GDP and above this, Fed interest rates, and the "real growth rate" which might be quantified by some consumption indicator. As you'll see, at any moment, the cross section of the Debtberg moves to the right, into the future, and those incomplete lines represent how the Debtberg might collapse with a recession, or worse. Next slide?

This next slide says that over some period of time, as the probability of mortgages actually being paid off declines, the bottom time scaled line pulls in. This in turn causes only the top of the Debtberg to become heavier, and it then rolls over in a crash-like way. And example of this kind of crash might be the Japanese markets since the mid 1980's when land in downtown Tokyo was going for over a million dollars per square foot. Any damn fool could see that was not going to be paid off - ever in most cases. This is where the confidence game is played out. Along this axis, we lie to ourselves and go to denial. When finally, repudiation of the debt becomes a conscious thought, the top heavy Debtberg is toast - and we are where we are in Japan today. Depression. Next slide?

Now, given that you see how a Debtberg behaves, here's the point I was making last week with the wing analogy. As you'll see in the top sketch, the economy was flying along just fine. We were whizzing through time in an economy with a lot of lift. But now, as we are past the peak of the Debtberg, you can see how the Fed has lowered rates, and the base of mortgages has continued to expand. Bet me that I can't find a 40 year mortgage going forward from here. As trends continue, the economy begins to fly upside down and counter intuitive moves such as restructuring debt to twice as many years, is the only way out of the crisis.

Social Security is a great example of this, but there are others. What these restructures do is change "lift" from positive to negative. However, the control surfaces remain the same (rates) and small in proportion to the debtberg'a dimensions. You almost never hear about someone having a "mortgage burning party" these days. What's the incentive to pay it off? You get a tax break, better cash flow, hell, you'd have to be a fool to pay off a mortgage. Until, of course, the wing that's now flying upside down, crashes. Next slide?

This last slide shows what we've been saying for several weeks in one way or another: As we go forward from here, lowering interest rates to the point that consumption will be so stimulated that the "airfoil" will continue to provide positive lift, can not be done by simply lowering interest rates. This is what we call the problem of a "to small control surfaces". The only way to get GDP up fast enough to rebuild lift, is for a huge increase in consumer spending to occur,. That can't be done without some kind of government subsidies. Ergo, the short-term route will be a U.S. tax cut. But will it be enough? Probably not, but it may slow the rate of change. Even so, the odds continue to be high that Bush II will replay Hoover.

Ehor's Calculations: The Math Behind the Mazurok-Ure Correlation.

I asked Ehor Mazurok if I could share some of our discussion and the math behind this view we've developed - and he was kind enough to agree:

"Generally, in beginning any kind of analysis,
the first thing I try to do is find some mathematical model or equation to
work with. In my experience this is the best way to get started.   With an
equation it allows one to explore several variables, see what the results
say, and compare them with reality.   If there is a match, then success.
No match means one has to go back to the drawing board.
On the other hand if an equation is not found, then comes the next step in
the process, a relatively obscure technique called "Piecewise Linear
Approximation of A Non-Linear Function." Here, a non-linear relationship
is fitted with straight line segments that approximate the function.
Those linear segments are examined if some relationship exists between
them. If a relationship does exist, then some vector or table is
constructed that points to the direction and magnitude these short line
segments will take.   Prediction is made by using the vector or table to
extrapolate where the next segments will go. If this method fails, then
the next step is charting, heads-and-shoulders diagrams, wave analysis,
astrology, or whatever works that allows some reasonable prediction to be
made.
>From the previous work emailed, you probably have surmised, my method
is based on the Linear approximation technique. The advantage of
describing linear trend by equations is it removes a lot of subjectivity.
If I give 100 people the equation y = m*x+k, then for a given value of
"X" about 90% of them will give you the correct value for "Y". (The other
10% will make some silly addition or multiplication mistake!) On the other
hand, if you give someone a chart, and ask them to identify head and
shoulders, cycles or trends, out of the same 100 people, you will get about
16% of the people whose results will be almost the same, the rest will be
all over the place. Its a question of reliability. Every method has a
certain track record which I call reliability. A formulized approach is
the most reliable, the ad hoc method is the least reliable.
Now back to the M-U correlation. As a starting point, I decided to use the
mortgage equation as a model to play with. The mortgage equation consists
of a DEBT lets call it "A" and a recurring payment schedule called "P".
Instead of thinking of this as a repayment schedule, assume "P" stands for
the on-going production that has to pay off debt "A". I am not sure how
that can be quantified at this time, because GNP (Gross National Product),
or NNP(National Net Product), or NNI (National Net Income) in my opinion
does not give a true measure of production paying off the debt.    The
quality of these statistics often get diluted by manipulations in the supply
of money. But this can be left for later consideration.
The classic mortgage formula is:
P = A*i*(1+i)^n / [(1+i)^n -1]
where   A is Debt,   P is production, " i" is the interest rate and "n" is
the production period, which for this formulation will be in years.   Since
Debt is of concern here, for ease of analysis the equation is reshuffled
to:
A = P*[(1+i)^n - 1] / i*(1+i)^n
cleaning it up a bit,
A = (P/i)*[(1+i)^n - 1] / (1+i)^n
Assume we increase "n" to infinity, we find that,
A = (P/i)because the fraction [(1+i)^n - 1] / (1+i)^n goes to "1" as "n"
gets very large.
This result will be used in later substitutions.
Using this equation, the next step is to determine what happens to Debt as
the term "n" is increased? To get an idea what happens here, take the
derivative of "A" with respect to term "n".
so     d(A)/dn = A*n / [i*(1+i)^(n+1)]

Figure 1 shows a graph of this function from 1 to 150 years....

© 2001 Mazurok-Ure Correlation

Very interesting result! It shows a peak or maximum at about 31+ years!
What it says that for a given production "P", (eliminated from the
equation by relationship A = (P/i)), maximum accommodation of debt occurs
for some value of "n". (around 31+ years).
So up to 31+ years, increasing the repayment schedule allows one to pile on
more debt. After 31+, the incremental amount of debt starts to decrease!
Already this shows the Kondratieff to be a debt cycle. Up to 31 years it
is worthwhile for business to increase debt, on the UPCYCLE, however, after
31 years, the law of diminishing returns starts to set in and the amount of
debt that can be undertaken diminishes. Somewhere after 31+ years on the
DOWNCYCLE, the economy finally says enough is enough and the music stops!
I don't think its complete heresy to guess the Length of UPCYCLE = Length
of DOWNCYCLE, so this would suggest, the Kondratieff is about 62+ years
long for the interest rate of 3%.
By using marginal utility as an analytical tool, the exact location of the
peak can be found. Marginal utility consists of comparing the next
increment with the last one and looking at the difference. When that
difference is zero, a turning point, a peak or bottom is reached.
Off we go! For years "n" we have A*n / [i*(1+i)^(n+1)]
and for "n+1" the next increment A*(n+1) / [i*(1+i)^(n+2)]
Subtracting the two to get that incremental increase, and equating it to
zero to get that peak!
Canceling everything results in   n = (n+1)/(1+i)

or     n = 1/(i)

an extremely simplistic result, but nevertheless it is true.
In our example i=.03
then           n = 1/.03 = 33.333 years.
Performing some quick sensitivity analysis produces very interesting
results. For example, increasing the interest rate to 20% reduces not only
the time to the PEAK, but the amount of incremental DEBT that can be
accommodated. For A= 10,000, maximum incremental amount shown in Figure 2
is about 83,000.

© 2001 Mazurok-Ure Correlation

In Figure 1, when the interest rate was 3%, the amount of
DEBT that could be managed is 4 million. If the interest rate is further
decreased, as shown in Figure 3, then the incremental amount of debt is
increased to 9 million and the maximum moves to 50 years away!!!!

© 2001 Mazurok-Ure Correlation

So in my opinion George, this confirms your assertion how debt behaves with
a change in interest rates, is right on the money. The mortgage equation
in my opinion is very good model to work with.
All of this has some very insidious consequences!!!! From Figures 1 and 3,
irrespective of the term of DEBT, the interest rate controls where the PEAK
occurs. ( hence formula n = 1/(i)).   The real problem is that before the
PEAK, increases in the term of debt, or rollover, results in an economy
that can manage more DEBT. However, after the PEAK, successive rollovers,
or increases in term of debt, result in diminishing returns. The economy
is able to manage lesser amounts of debt.    In other words, whatever works
before the PEAK, the opposite works after the PEAK.
So lets assume the economy is at a PEAK, and faced with diminishing returns
( a recession ), Mr. Greenspan decides to drop the interest rate. This
moves the PEAK further out, the economy can take on more debt, and happy
days are here again as the economy just keeps on trucking! How long can
this go on? Obviously not forever, there is a limit how far into the
future this peak can be pushed. Sooner or later one has to go past the
PEAK, when the fun really begins. More outstanding debt than money means
higher interest rates, but higher interest rates mean the economy
accommodates less debt. The only answer is massive defaults on DEBT
resulting in a big recession, or print money like crazy and have a nice
hyperinflation.
If I am right, Mr. Greenspan has made a serious miscalculation this time
around. Around year 1997, I believe debt peaked, and in dropping the
interest rate, Mr. Greenspan cause that PEAK to move forward so the music
wouldn't stop. When Mr. Greenspan raised the rate, in a sense he pushed
the economy over the PEAK where it could not sustain the debt an longer, so
LTCM collapse was the result. Again, Mr. Greenspan is attempting to
forestall the PEAK with a rate reduction, but debt has already been pumped
up to such an extent, that if there is an increase in interest rates, the
economy whizzes past the PEAK with lightning speed, and another round of
defaults precipitates.
The difficulty in all of this, is figuring out what to do. At the start of
the K-Wave, playing with interest rates is no big deal, increasing them,
reduces debt, while dropping the rate allows more debt to be piled on.
Hence stimulative policies work miracles. However, once passed the PEAK,
things work in opposite direction. Dropping the interest rate results in
the opposite effect, that is decreasing the amount of debt the economy can
manage, and bringing a recession instead of a recovery, the opposite
desired effect. If there is anything that will cause the FED to lose
control, this I believe will really be IT!!!

The only thing I can think of adding to Ehor's math, is the notion that the bottom line of the Debtberg model (the one with years and mortgage rates) could better present today's economy if the time scale in years is spaced according to the underlying velocity of money. But perhaps we should leave that for when we get invited (by a think tank or brokerage house somewhere?) to go pencil out more tradable details....

Be sure to Visit:

Treasury's Office of the Public Debt (Click here to get there): These folks in Treasury are one of the last bastions of candor in the federal government!

Back to main page