In the previous article we’ve established the existence of the extra-long debt cycle, which consists of the accumulation and unwinding phases. The next step is to evaluate the trends in monetary aggregates and interest rates.
For reference, let’s repeat the Great Debt Cycle, i.e. debt/GDP cycle chart that we saw before:
This credit cycle is called Kodratieff Wave, named after Russian scientist Kondratieff.
1990 anomaly
For futher reasoning, it is important to note that from the cyclical and absolute point of view it looks like the 1990 was a moment when the debt cycle was about to reverse and start unwinding, but the destruction of communism and the improved productivity from Internet and computers (see also “funny jobs“) prevented the unwinding and instead the cycle continued to rise into unprecedented levels.
The anomaly of 1990 is also quite visible on the monetary aggregates charts, which is why I have to mention it.
Monetary aggregates
The M1 money stock (currency plus demand deposits) Y/Y growth chart:
From the first chart we see that the current debt accumulation phase started in 1945 and is still intact as in mid-2008, but the M1 money supply growth shows a different picture. It was accelerating from 1945 to 1985, but since then is experiencing the deceleration and already experienced several periods of outright contraction. The anomaly of 1990 manifested itself by resumed accumulation of M1 a decade later (it seems that money aggregates are trailing, not coincident indicators). The last contraction started in mid-2006 and is still ongoing.
The M2 is M1 + small time deposits (less than $100,000), savings deposits, and non-institutional money-market funds. Here’s Y/Y growth chart:
On this chart the anomaly of 1990 is visible very well. The M2 money supply growth accelerated from 1945 into early 1980s and then demonstrated a perfect, textbook, deceleration. But the life is not a textbook and M2 accelerated again into 2001, but since then the trend is down again. Looking at the charts we are likely to experience a further deceleration of M2.
In general I don’t like M3, because it includes institutional money market but doesn’t include other [formerly] quasi-cash instruments like auction-rate bonds or commercial paper. So instead of M3 I present the chart of M3+debt from nowandfutures.com:
Again, the most broad money supply growth is in downtrend since 1973. The growth nearly stalled during the “anomaly of 1990″, but resumed again. The blip in 2008 is probably explained by the drain of credit facilities that happened because many wise corporations rushed to take what they still can.
Looking at those charts we established that the credit accumulation phase of the Kondratieff Wave could be further divided by half with monetary aggregates growth accelerating at the beginning but decelerating later on.
Interest rates and inflation
The next chart represents the historic interest rates, including corporate bonds (reproduced by permission from Ned Davis Research):
On this chart you can see that interest rates were on the rise during the first 37 years of the credit expansion cycle (1945 to 1982), then declined from 1982 till the end of this chart (2008). It is easy to see that this spike had the inflationary nature because the spread between instruments with different credit rating is in fact quite small. That suggest not the particular tightness of the lending nor any kind of risk aversion but a plain simple premium for the declining dollar.
Another visible spike happened in 1932, during the credit unwinding phase of the previous cycle (i.e. before 1945). But the spread was not just huge but record at that time. It suggests that the spike in interest rates is produced by the general shortage of money and risk aversion, when the more risky borrowers had to offer a huge spread to borrow. Please take into account that BAA corporate bonds are not junk, this is investment grade paper - and it still reached 11% in 1932. It’s also quite natural that BAA bonds spiked during every recession or mid-cycle slowdown (1994, 2006).
Credit growth, inflation and monetary base
From the charts posted it is quite visible that the debt expansion accelerated after inflation peaked in 1982. At the same time the monetary base growth slowed.
It is pretty hard to carefully formulate why that happened. I think that during the period of the secular inflation (1945-1982) the growth of the long-term debt was handicapped by the inflationary premium. The fixed income market was in predominantly bear market, which made all existing instruments to perform not quite well. But the push from the real economy to grow was quite strong, the world was recovering from the WWII and the infrastructure build-up all around the world was producing a strong demand for anything useful. The expanding economy produced liquidity, because the amount of liquidity has to expand when the economy gets bigger. If long-term fixed income instruments were not quite desirable by the market the demand shifted into short-term instruments, which are closely related to monetary base.
In opposite, when fixed income finally entered the secular bull market (after 1982) the real economy started to shift from short-term to long-term instruments. Long-term bond boom is removing some heat from short-term debt markets, effectively cooling them off. The slowdown at the short-term debt market is removing the need of accelerated buildup of monetary base (M1 and M2). But don’t be confused, there is no slowdown at the debt market at all, it is just a shift in maturity preference that is producing the described effect.
Moreover, I think that when the credit instruments are in the bull market they are somewhat treated as money. In many cases all temporarily available funds can be parked in bonds and thus money are freed to move elsewhere. What I mean is that bonds funds and money market funds can replace deposits, i.e. act as money.
The credit unwinding phase
On the chart above the credit unwinding phase (1929 to 1945) is manifested by the small decline of the yields (1929 to 1931), then the sharp spike in interest rates (1931 to 1932) and then the gradual decline of interest rates all the way till the new credit cycle starts.
Apparently initially the money are moving from stocks to treasury and AAA bonds, then the real credit crunch hits and the price of money is moving up, manifested by the sharp increase of all rates. The 1-2 years of high yields is sufficient to throw the financial system into deflation, which moves the rates down. Eventually the Treasury bills settled around 0.25% rate and long-term Treasuries in the 2.3% area.
In the Japanese credit unwinding cycle of 1990-2007 the rates settled even lower, very close to 0% (called ZIRP, or zero interest rates policy). As the US debt/GDP ratio in 2008is surpassing the credit bubble of 1920s it is possible to guess that the interest rates at the end of the cycle will resemble Japanese rates rather than 1930s in the US.
Here you can see that the “credit crunch” spike of the 10Y Japanese government bond rates to over 8% in 1990 was followed by decline into sub-1% area in 2003. Maybe just to scare Mr. Greenspan.
For reference - the Japanese stock market:
Comparing those two charts above is a very important exercise. As you can see the interest rates peaked in late 1990, when the Japanese stock market was already in a full crash mode. That tells us that the final peak in interest rates happened not because the economy was “too hot”, but because of the credit crunch being in full force
- In the following articles I will describe the seasons of K-wave. Please stay tuned
To subscribe click 
17 Comments
Isn’t adding debt to M3 double counting? Debt is the flip side of the components of money supply. Money funds and banks hold those debts as assets. Money supply data measures the money liability.
>>> Isn’t adding debt to M3 double counting? Debt is the flip side of the components of money supply
Agree. I don’t know what to do with M3. It includes cash-like short-term credit, which is already someone else money and could be lent to another party, again. There is a lot of double-counting in M3.
I would suggest to ignore M3 as it’s IMHO bogus, with or without debt, but I’ll keep this chart above in my post just for reference. It’s a nice chart.
There are so many websites that reconstruct M3 to show that it’s growing. My answer would be always “so what?”
Good article.
I like multi-decade charts.
Money should also include short term treasury securities. The fed has released $400 billion of short term treasury securities in exchange for toilet paper as collateral. This is highly inflationary.
>>> The fed has released $400 billion of short term treasury securities in exchange for toilet paper
Yep, that’s why rates are going up. See Japan chart above with rates at 8.5% in 1990
#2.
Completely agree. For the sake of looking at the money trend data on a consistent basis I stick with MZM, but of course my contention has been that the data grossly overstate the amount of money that really exists and can actually be tapped. My gut feeling is that about a third of the reported data is backed by fictitious capital. If they ever try to get their money out of the MMFs on balance instead of pouring it in all the time, they’d find out real quick that it isn’t there. A ponzi scheme can continue as long as investors continue to invest ever greater sums in it. So far, that’s been the case with MMFs.
#5-
Where’s the $400 billion come from? TAF is $150 billion. PDCF is next to nothing now, and TSLF is down to $64 billion. The total swapped is now around $215 billion, plus another $13 billion lent at the Discount Window.
And while some of the collateral is toilet paper, not all of it is. The Fed has the right to exercise the death penalty if the collateral goes bad, so I’m pretty sure that most institutions aren’t sending their worst crap to the Fed.
Besides, I don’t know that Treasuries aren’t toilet paper, the way government debt is mushrooming, and revenues are falling ever shorter of the ability to pay the bills and service the debt. The day is coming when US government debt IS toilet paper.
#7 - I see. $215 bln is not a joke, one day the market will choke when so much new T-bills are thrown in its face.
At this point AAA Treasuries trade differently from AAA corporate. I think first step will be that the spread narrows, if any AAA corps will still be around at that time
Released treasuries?
Since when is releasing treasuries talked about in the same way as “releasing” cash? The old vernacular is Fed buying treasuries and giving cash to banks, grows the money supply.
No, it’s not about the treasuries.
It is more likely the GSEs effect on money balances that you can say is growing money, not the Fed.
Roxy, just wanted you to know I am following your series and awaiting with interest your discussion of transition to K winter.
I’m curious as to the source of T-bill rates during the 1929-32 period. St. Louis Fred only has them beginning in 1934. One of the items I am most interested in is the shape of the yield curve during that time. We know the classic inverted yield curve preceded the 1929 crash. If the yield curve remained positive (albeit much higher than the inflation rate) thereafter, that’s obviously an interesting point.
The chart of interest rates came from NDR.com - Ned Davis Research.
They gave me the original PDF file and permission to reproduce it.
Yep, there is a spike of Treasury interest rates in 1932 and I will cover this in the future article.
>>> Released treasuries? Since when is releasing treasuries talked about in the same way as “releasing” cash?
I agree with you and I think I did not make it clear above.
Dumping treasuries by Fed is not inflationary at all, but it drives interest rates up because the market is not infinite and swallowing another $200 bln, especially when charts are not looking good, is damaging the price
NDD, please look at my update of Japanese charts above, that will be interesting for you.
Roxy, the only source I’ve been able to find as to short term treasury rates during the 1929-32 period is here: http://www.economagic.com/em-cgi/data.exe/fedbog/tbaa3m
Did NDR give you access to the data they used to generate the chart? Is the above 3 month T-bill the data set they used? It’s not so much the spike that got me as the positively sloped yield curve throughout.
The the JPN update, like the US, the bond blowout (due to a wave of defaults in the US) didn’t happen until well into the bust.
Thanks
Yes, thanks for the update. Way back when I first started getting into investing and watching the markets, I read an ancient text in which the historical readings for the DJ Bond Average were given. That average accurately presaged the 1929 crash, and held up very well for a year or two thereafter until all the bond defaults hit the fan. Seems like Japan had a similar experience.
I am quite interested in the T-bill data NDR used in constructing the 1929-33 part of their chart, which appears to show a positively sloped yield curve for almost the entire duration. The only data I’ve been able to find is here. Any way you can confirm/post the data they used?
Your spam filter is killing me!
Twice I’ve tried to add a comment including a link, and twice this site has vaporized it. So, here goes the more laborious and sneaky way….
I am really interested to find out what data series NDR used to calculate T-bill yields back in the 1929-33 period. The only data set I’ve been able to find (sigh … here goes the laborious, sneaky part) is at
www dot economagic dot com slawh em-cgi slach data dot exe slash fedbog slaxh tbaa3m
(go ahead spamkiller! Let’s see if you recognize that!
I say that in particular because it appears the yield curve was positively sloped after Oct. 1929 throughout the entire period.
At one point I pored thru an ancient text of DJ Averages. The DJ Bond Average gave a perfect signal in early 1929. Bonds didn’t really blow up until there were massive defaults in about 1931.
NDD, they gave me just that chart. Yes, positive slope all the time.
Yield curve is an implied forward interest, made by the market participants. So I think there is no magic in that except those guys are usually smart, unlike stocks guys
They just got it wrong all the way down.
2 Trackbacks/Pingbacks
[…] previous two posts on the Kondratiev Wave discussed the debt cycle and monetary base as the basis of the K-wave cycle. The goal of those articles is to show that there is nothing […]
[…] Monetary base and interest rates […]
Post a Comment