After the introduction of the seasons of the Kondratyev wave we’ve proved the existence of the extra-long secular cycles that manifest themselves as the periodic swings in the relative debt level, monetary base, and stock market. But the more complicated task is to actually analyze the causation of those cycles.
Even before getting into macroeconomics it is possible to make some basic observations and hypotheses using the simple theory of feedback loops.
For reference - here is the list of previous articles of this cycle:
The definition of the feedback loop
The feedback loop is a process where some proportion of the output is fed back into the input. The feedback loops can be classified as positive or negative depending on whether the loopback signal is reinforcing, amplifying the input or muting, stabilize the input. Here are the definitions we’ll use:
- Positive feedback loop - the feedback influence is reinforcing the main trend (positive has nothing to do with “good”)
- Negative feedback loop - the feedback influence is working against the main trend
- Vicious cycle - positive feedback loop producing undesirable economic results
- Virtuous cycle - positive feedback loop producing desirable economic results
Here I’m purposely not giving any definition of “desirable” or “good” economic results, because sometimes similar economic trends could be “good” or “bad” depending on conditions. For example, it is very good to see some inflation when economy is experiencing a protracted deflation, but elevated inflation is very undesirable. It is better to define a particular “good” in particular moment of time.
The big picture
Let’s get back to our system. We know that the whole K-wave is oscillating, i.e. returning into original point every 50-70 years. That means that there is a certain negative feedback mechanism that return the pendilum back when it swings too far. But it can’t be as simple as that, because in that case the swings would be fast and short.
It means that besides the big negative feedback loop that can only repeat itself once per 60 years there are positive feedback loops that are able to push the system in one direction for as long as 30 years. The power of those positive feedbacks can be proven by the fact that there are some small oscillations as well - the economy also has the basic boom-recession cycle that is oscillating more or less every 4 years. Please note that sometimes the recession is avoided (like in 1995 and 2006) and this is called the “mid-cycle slowdown” or “goldilocks economy”, i.e. it makes the impression that the cycle became longer. Those short oscillations are obviously produced by various negative feedback loops.
So the existance of the negative feedback loops of short duration (4 years) does not break the 30-year swings in one direction, which proves that the positive feedback that is making this giant move is in fact very powerful and may not be broken simply by the swings in the business cycle.
The K-oscillators and feedback loops
Let break the K-wave into seasons and repeat the simple observations collected in previous parts of this cycle:
| Season: | Spring | Summer | Autumn | Winter |
|---|---|---|---|---|
| Debt cycle | Growing | Growing | Growing | Unwinding |
| Monetary base | Growth accelerating | Growth accelerating | Growth decelerating | Declining |
| Inflation | Inflation | Hyperinflation | Disinflation | Deflation |
| Stock market | Bull | Bear | Bull | Bear |
What we can say looking at this table before applying the feedback loop theory? First thing to say is that the number of expected feedback loops is at least three or four or even five for every row of this table, i.e. we should probably expect from 10 to 20 of main feedback loops (and we can immediately assume that there are many more secondary feedback loops).
For example, the Debt cycle is experiencing three seasons of growth and one season of unwinding. The expected feedback loops would be:
- The virtuous cycle that is producing the [desirable, lets not get into details] growth of credit during Spring and Summer
- The vicious cycle #1 that is producing the [undesirable] growth of credit during the Autumn
- The negative feedback loop #1 that is making the further growth of credit impossible during the late Autumn and eventually turns the Autumn into Winter
- The vicious cycle #2 that is producing the very painful unwinding of credit during Winter
- The negative feedback loop #2 that is making the production of new credit possible at the end of Winter, eventually turning it back to Spring
Here we can see that we started from possible model and produced the list of possible feedback loops even without getting into any details on how, when and why the credit is growing or unwinding. The modeling approach is by itself a feedback loop. First we use our intuition and knowledge of other fields to produce the simple model, then we try to fill the model with details and test it with experiment. If model is passing the experiment in general the additional details produced by experiment are fed back into model and then the improved model is tested with another experiment.
Example: the feedback loop model and ordinary (Kitchin) business cycle
Here I want to use the feedback loop approach to model the regular 4-year (Kitchin) business cycle. This exercise has very little to do with Kondratieff wave but is relatively simple and could be used just to demonstrate the method.
Let assume that the Kitchin cycle consists of 3 years of boom and then 1 year of recession/sluggish growth. Let start from the first year of growth and just enumerate all the feedback loops:
- Virtuous cycle - low interest rates triggered the investment community search for yield and the struggling corporate world that just survived the recession is offering the favorable investment conditions
- Investors are taking on more risk
- Good business plans get funded
- The unemployment decreases
- The consumer purchasing power is increasing
- The profits are rising
- The stock market goes up (feeds back to consumer purchasing power)
- The investors are rewarded (feeds back to investors taking more risk)
- Negative feedback loop #1
- The short term credit growth is feeding the inflation
- Velocity of money is increasing, feeding into inflation as well
- Eventually the full employment is feeding into inflation on its turn
- Inflation is growing, reducing corporate profits
- The cost of credit is rising
- Sometimes the business plan does not warrant the cost of credit - i.e. the business borrows more then it should
- It’s hard to retain good workers
- Negative feedback loop #2
- High rewards are making investment more reckless
- The stock market growth is producing the “wealth effect”
- Consumers are spending more then they should
- Businesses are spending more then they should
- A lot of money are wasted on something that is not producing further returns, bad projects are funded, debt is loaded
- Vicious cycle - when the negative feedback loops are finally weighting on the growth and the recession starts
- The stock market declines
- The credit becomes more expensive
- Bankruptcies are on the rise
- Investors are pulling money out
- Its very hard to get a project funded
- Unemployment is increasing
- Consumption is declining
- Negative feedback loop #3 - when economy is in recession, there are forces that are seeding the next recovery
- Consumption is declining but conservative investment is increasing
- Bankruptcies of inefficient firms are clearing the field for more innovative and prudent firms
- The inflation is declining
- Low interest rates are forcing investors to search for more risk/yield
- Venture capital may get into interesting ventures on very favorable conditions
- The cost of labor is stable and its easy to retain good workers
- The decline of profits is slowing down
- The stock market is offering very good investment opportunities
The basic knowledge of the economy is suggesting that the list above is very incomplete. It is possible to list many more feedback loops, but that was done for illustrative purposes.
Bifurcation points and Season overlapping (Indian Summer)
Some basic observation lead us to conclude that the various phases of the regular (Kitchin) business cycle can mute or amplify the main trend of the current Kondratieff season. It almost feels like an Indian Summer, i.e. the previous season may briefly return.
Another term to introduce is a Bifurcation point. It’s the moment when the remainders of the previous season break with a loud crack and there is no more doubt that the next season already started.
The example of season overlapping and bifurcation point is the economic history of United States in 1990-2007. The Autumn prevailed from 1982 and first wave of Winter gave some frost in 1990, then again in 1998 and 2000. But the Autumn returned again and again. Only when the securitized mortgage market collapsed in 2007 (the Bifurcation point) it became clear that the Winter finally came. You can say that the Winter started in 2000 and the 2003-2007 period was the overlapping Autumn.
…
Those are the basics of the theory that can explain the existence of ultra-long economic ccycles. The task to apply the method of feedback loops to K-wave will be quite complicated and I’m also hoping for some feedback from the readers.
-
To subscribe click
13 Comments
Andy,
Your article is over my head, but I think you are saying we have experienced all of the above both from a long-term economic up-cycle of say 50 to 70 years and we have also just finished a shorter-term economic up-cycle that began in 2000 or 2001?
So it appears that several longer term up cycles and shorter term up cycles are converging to generate the next economic down-cycle.
If correct this would also line up with certain long term demographic cycles that will reinforce a down turn.
If one is really into cycles, then it is possible that we may also be lining up with a pandemic cycle, which should start during the winter of 2009.
Wars are also on a cycle.
To be clear it is essential to note that Kondratieff spoke of supply/demand driven rising/rapidly rising/falling/rapidly falling price rather than “inflation/hyperinflation/disinflation/deflation” which are terms that describe devaluation/revaluation of the medium of exchange, not price movement.
Ever read anything about Martin Armstrong, and his Economic Confidence Model?
A cycle can appear based on a inverted feedback with a delay. This is an oscillator. The inertia of the system helps define the delay, and hence the cycle period.
Don’t confuse this with a periodic forcing function. If the President orders the Fed to have a roaring economy every four years, you get a four-year cycle without any need for the underlying system to have any resonances with a four-year period.
I’ve been subscribing to Tim Wood’s newsletter for about a year. It took me several months to get even a basic understanding of cycles and I still have a lot to learn. One thing I do know though and that is the perilous situation brought on by an extreme left-translated four year cycle. Think GD, only worse.
http://www.cyclesman.com/Articles.htm
Here, no, don’t know Martin Armstrong. What’s a link for starters?
Plantagenet, agree on both points. Very interesting, thanks.
I understand and agree with the theories you describe, however I realized recently I have never seen a debt to GDP graph of Japan over the last 30 years or longer, despite the fact that Japan is often used as an example of unsustainable credit growth (entering Kondratiev winter after 1989).
Does this data not exist publicly? You included other graphs on Japan (interest rates, stock market) in this series, and the US debt to GDP graphs spanning before the Great Depression through today have been widely published… but the Japanese equivalent is strangely elusive, even when I search. It seems it would provide valuable further evidence in comparisons with the US/UK/etc today.
Thanks for this interesting blog and series of posts.
A simple explanation for the 60-70 duration of K-waves is that this is typical lifespan of economic participants. The great grandparents screw up and cause a depression. The grandparents live through the depression, clean up the mess, and put safeguards in place to prevent future depressions. The parents don’t fully understand the safeguards but maintain them since the grandparents pounded into them that the safeguards are necessary or very bad things happen. The children come along and disassemble all safeguards since they are obviously obsolete in this new era.
The children then get to relearn the hard way all the lessons that their grandparents learned through bitter experience.
Wash, repeat cycle.
Andy,
You can find some smatterings of stuff if you dig around with Google. It is worth your time if you’re interested in cycles. Check this one out:
http://princetoneconomics.blogspot.com/2006/06/economic-confidence-model.html
The last major turning point in his model was Mar 22, 08. The next major turning point, and it is a big one as it is an inflection point, is Apr 19, 09. Downhill from there till Jun 13, 11. Around the time I think the housing market bottoms, but maybe I’m too pessimistic.
Armstrong links are very interesting, however the information on his models is insufficient.
As of timing of the K-wave I think the duration of every season can fluctuate in wide margin, making the length of the whole cycle unpredictable.
For example, the Autumn of 1920s was twice shorter than Autumn of 1980-1990s. 10 years difference just in one season! And this is in line with my analysis - every cycle is so much different from the previous that you must not apply the same ruler. That could make models based mostly on timing incorrect, because they will try to torture the data to fit their timings.
FWIW: Markets always swing like a pendulum. Spring and Fall are the crossing points since the pendulum must pass a distance between the two ends.
The issue I see with K-wave analysis, is that is impossible to predict when the pendulum will swing. Technology, demographics, and geopolitics all effect the economy. We should already be in a deep winter but we are not, and we are well beyond a standard deviation. This can be attributed to technology which really took off since the 1980s with advanced computing technology. Energy was dirt cheap since the mid 1980s until about 2004. These two forces have impacted the pendulum’s swing. A long periodic cycle cannot be use to reliable predict when economic changes occur.
K-wave analysis is a lot like earthquake predicting. if we take the average time between quakes, when can roughly guess when the next quake will hit. But there is no way to pinpoint when the next quake will occur. It could hit tommorow or 50 years from now.
OT: Freddie and Fanny got hammered today. Both stocks down more than 15% each. I believe part of GSE capitialization comes from stock price. A steep fall in stock prices might be taking a bite out of their capital reserves, and the falling stock price will also maket it more difficult for them to raise capital.
Considering that about 90% of new loans and refis go through the GSE (since most of the large banks already have cash problems and capitalize loans through the GSEs). This could be a big deal. If the GSEs credit machines are shutdown, is it game over for credit and the US economy?
I recall on CNBC last night that housing bill has been stalled. Apparently Nancy Pelosi wants to focus on passing another stimulus package first. This may be related to Bush’s threat to veto the housing Bill. Bush may be onboard for antoher stimulous bill. Of course if credit dries up, no stimulus is going to prevent an economic implosion.
>>> We should already be in a deep winter but we are not, and we are well beyond a standard deviation
Agree. If we talk numbers we experienced a big, almost unprecedented upswing in productivity, which possibly was one of the factors that Autumn extended into 25 years. What’s good, the productivity is still high in the middle of the credit crunch. That is probably unprecedented. So is the debt/GDP.
>>> If the GSEs credit machines are shutdown, is it game over for credit and the US economy?
Even hard to imagine
>>> I recall on CNBC last night that housing bill has been stalled
Btw, Roubini mostly supported the housing bill. He said the bailout of many more banks could be more costly comparing to the cost of that bill.
One Trackback/Pingback
[…] The previous chapter of this cycle introduced the concept of Feedback loops and the Vicious Cycle. During the previous season, Spring, the government used to print money without big spikes of inflation. Once the transition to Summer happens (it’s a topic for a separate chapter), the money printing (monetizing of Treasury debt by central bank) triggers the vicious cycle: […]
Post a Comment