Time to draw them all together and see the full correlations. This is US-based analysis due to data availability.
The first chart (click to view larger) reveals historic spikes in US interest rates, 10 year treasury yields, MZM money velocity and US inflation (averaging official CPI and Shadowstats data) all within a 2 year period around the solar maximum (note the 1968 solar max was November and the 1979 solar maximum December, hence the 2 year boxes following; also note some of the measures have been scaled to share the same chart).
Stepping back further in time, the 1947 solar maximum was accompanied by a 1947 inflationary peak, followed by spikes in corporate bond yields by 1948 and treasury bond yields by 1950.
If the next solar maximum is ahead in Autumn 2013, then by history we should see spikes in rates, yields, velocity and inflation within around 2 years of each other and of the solar maximum. Is it different this time because the government has acted to surpress both interest rates and bond yields? With velocity correlating closely with bond yields, is an inflationary peak not going to happen this time? I believe it will happen, as the same surpression occurred in the 1940s and yet the spikes took place.
The second chart (click to view larger) adds in real commodities using the CRB index adjusted for inflation (and again scaled). Interestingly, real commodities behave very similarly to rates, yields, inflation and velocity – all moving together into peaks (orange boxes) and troughs (red boxes), over periods lasting around 3 years.
There is a general pattern of collective peaks around each sunspot peak, and additional collective peaks before solar mimima. I don’t yet understand why we see rallies leading into solar mimina, however they have historically set up the panics and crashes that occur at the solar minimum. Nonetheless, yields, commodities, velocity and inflation all acting together is suggestive of waves of ‘human exctitement’ that brings about speculating, buying and circulating money in the economy, or the opposite.
The third chart adds the real inflation-adjusted S&P500 and US demographics trends (middle to old and middle to young ratios combined) into the picture. Here we again see evidence of ‘human excitement’ correlating with sunspot peaks as some combination of real stocks, real commodities and inflation spike up around the solar maximum.
Demographic trends appear to be important for real stocks to peak, whilst commodities appear to behave opposite to demographics.
In summary, there appears to be a 4-way correlation between equities, sunspots, demographics and inflation, whilst there appears to be a 5-way correlation between rates, yields, velocity, inflation and real commodities. My solar-theory take on it is that the same phenomenon of human excitement (driven up and down by the solar cycle) translates into trends in buying, asset speculation and circulating money, hence the united correlations, whilst demographics (which also have a solar input: solar maximum peaks (and occasionally troughs) in births) additionally feed into equities due to investment/disinvestment in equities, relating to retirement.
I’d be interested in your thoughts on any of the correlations in the charts. I suspect there is more to be teased out.
42 thoughts on “Sunspots, Equities, Treasuries, Commodities, Inflation, Money Velocity, Interest Rates And Demographics”
Thanks again for sharing you insight.
Looking at demographic vs commodities it seems that by this solar maximum we would be more similar to 1969 then to 1980. It is just pure speculation but it maybe that 3×11 years cycle in commodities was occurring in 19 and 20th century and it got extended to 4×11 by now, due to longer life expectations. This would mean we have to wait for spike in commodities little bit longer.
Interesting, thanks Pawel. So you mean commodities would need another 11 years and this speculative peak would be the rally in equities we are seeing?
Our thought directions are mostly not our own but come from the group mind – which is itself a product of collective feeling at deeper levels. Cycles of variation in solar energy would effect these. That is why suddenly the mood changes and society moves in a new direction. Media then tries to explain these after the event.
Sorry but my thinking has nothing to do with any kind of collective mind:) It’s true though that the majority of people are conformists and they try hard all their lives to fit into the mass, not to stick out in any direction. Only some are independent to create, not to copy.
Thanks. Got any links on that group mind John?
It would be wrong to assume that demographics are responsible for economic upswings and downturns. They do exaggerate the upswing or downturn, and so a lot of the policy decisions made by governments and central banks aren’t always going to be the right course of action, unless demographics can support them. It would seem to be the rule of thumb that nations enjoy their own economic booms, and when the demographic window closes, the decisions made during the boom years dictate how the country can perform in the future. For Europe, those days were in the 1970s-1980s, and now we see low economic growth as a result of policy failure, worsened by population ageing.
Japanese policy makers assumed Japan could withstand high interest rates in 1990, believing the economy was rigid enough to handle it, but the population was starting to age and later on, Japanese governments started to see lower tax receipts and so this sparked two lost decades. South Korea has managed to enjoy economic prosperity recently, thanks to liberal economic policies and a growing middle class to keep tax receipts buoyant.
Much of Europe has failed to liberalize politically and economically, so when the demographic window closed, they found themselves consigned to decades of slower economic growth. The death of Baroness Thatcher highlights something very important. Thatcher made the most of the demographic sweet spot in the UK with her economic liberalism. Many mainland Europeans look to Thatcher with some admiration. She was the politician that mainlanders wish had reformed their economies.
Hi John. Just looking at the first chart, the correlation doesn’t look all that good. This is for a few reasons:
1- your 2 year time window is rather loosely defined. Normally a 2 year window “around” a solar peak would start 1 year before it and end 1 year after it. Here it looks like the yellow boxes are moved to catch the inflation peaks as much as possible. While there is nothing wrong with that, it means that effectively a 3 to 4 year time window is considered, within which the 2 year yellow box is then draw in with the benefit of hindsight.
2- marking peaks in a chart is also fairly subjective. If we gave this chart with the yellow boxes and solar cycles removed, asking people to mark the inflation “peaks” they see, then some people may mark only 1980-81. Others will also mark the smaller peaks like 2000. They will see about 7 peaks: 1969-70, 1974-75, 1980-81, 1989-90, 1994-95, 2000, 2006-07
Of these 7 peaks, only 4 were captured by the movable 2 year boxes around solar max, while 3 were missed.
Given an effective 4 year window within an 11 year cycle, one would normally expect almost 40% of the peaks of anything to fall within these windows. Then to find something that has captured 4 out of 7 is not a great “correlation” at all.
A loosely defined peak window also becomes a problem in real time trading, when we have to act without the benefit of hindsight. E.g. If we look at the 1980 peak period. Solar cycle peaked in Dec 1979, interest rates peaks around summer 1981, almost 2 years later. If we happen to get the same scenario now, then a late 2013 solar max would translate into interest rate peak by summer 2015. But if we get a repeat of the 1959-60 “peak” , then interest rates would remain low and continue to grind lower into 2015. So, one could be losing a lot of money by shorting bonds into 2015, only to discover that the ’59-60 scenario is in play.
Knowing that a peak can be expected within a 3-4 year time frame, is of little use if we don’t know whether that peak will come early or late in the period. Whenever considered time window is so big, any information it might contain is becoming very diluted. Just like we could say that the stock market makes major or minor peaks every 3 to 4 years, which is generally true, but that knowledge is of little or no practical use.
My priority is the practical application right now – namely is a combined rally in yields, inflation, real commodities and velocity ahead? Does a solar max produce it without fail and if so when in relation to the max? From the data, I suggest tentatively that it does, and that it should occur within 2 years from now (the bias in the peaks would be after the solar max). If that were so, then I can position for it and so it’s useful. But it’s tentative and historic occurences are limited – it’s an experiment in progress. For me, the bottom line is which tool gives me the best probability of how developments in asset markets will unfold this year and next, and it’s solar cycles. There’s a looseness in the timings that means I need to take it step by step and cross reference with other developments and indicators, and that’s my approach.
Here is the longer term inflation chart: https://solarcycles.files.wordpress.com/2012/02/17sep18.png
The question to an impartial observer should be is there an inflation spike at each solar max – as that’s what I want to know rather than what happens inbetween – and I suggest there clearly is. But we first need increased evidence for the solar max being ahead this year rather than back in early 2012.
Perhaps I need to break down the above charts so they are less busy and it can be seen clearer the performance of each data set.
OK, but it is in the practical application “right now” that all these problems will appear and potentially put a trader on the wrong foot with his expectation based on solar max. To draw these yellow boxes with the benefit of hindsight is easy, to do it in real time will be much more difficult. For example if early 2012 is retained as the peak of the current solar cycle, then you will be drawing your yellow box over 2011-2012. But there is no way to know it right now. So how are we going to trade right now based on solar cycle information that will not become completely clear until one or two years from now? Sure, one can make educated guesses whether solar peak is behind us or yet to come. But even NASA has been struggling at that, so what makes us think that we will know solar max when it is upon us?
Apart from the problem of timing the solar max and thus the proper positioning of the yellow box, we can also take a better look at how practical trading of this system would work in real time.
A person who wants to trade your system would probably trade the 10Y treasury futures. The 1959-60 solar peak didn’t produce much of a move. In the 1970 and 1980 yellow boxes a trader would likely have made money by betting on higher rates once he guessed that the solar peak was in.
In 1989-90 the 10Y treasuries green line peaked well before the yellow box, so a trader expecting rates to rise even more in ’89-90 would have lost money.
In 2000 the two year box would also include 2001. Same problem there, giving the rather small rise in interest rates in 2000, a trader would probably have waited for an even higher interest rate peak in the next two years, only to see rates go the other way, again losing pretty badly.
One might suggest that exact timing of the solar peak is not a big problem and hope that the trade will still work out even if 2012 retains the SSN max, thinking that a secondary later peak would create some rise and second peak in rates as well. But 2002 already proved otherwise. There was a secondary solar peak, coming close the the 2000 ssn, but there was no second peak in interest rates or the stock market. In fact 2002 was a major low for rates and stocks.
It isn’t sufficient to draw charts with yellow boxes that match some peaks well, and then think that one would have benefited from that information. To trade it in real time would have been much more challenging than this chart suggests, and not rarely waiting for higher rates in the two years after solar max would have kept a trader on the wrong side of the market for a year or two.
Sure, one can try to take in help from other indicators or sentiment readings, but then the success or failure in trading will come mostly from these indicators rather than from the solar cycle consideration. To expect an interest rate peak in some 2-3 year time window “around” a solar peak that can often only be timed properly two years after that peak happened, that is just too vague to have practical real time use. First one has to guesstimate the time of solar max fairly well, which is already hard in itself. And then one has to guess properly whether the expected interest rate peak is coming early or late in the estimated 2 year yellow box. That’s a lot of things that have to right for this system to “work”. One can go on sitting on losing treasury short positions only to discover after one year that the yellow box is not where one has expected it to be, or one might discover that the yellow box is OK, but treasuries had already bottomed out a few months before the estimated 2Y yellow box started.
By the way, given the known correlation between inflation and interest rates, you could simplify the chart by calculating the average of your 4 data sets and just show it in one line comparing it to the 1/12 sunspot data. It then also becomes more easy to calculate the actual correlation coefficient between these two data sets. I think you are too optimistic about the actual correlation in this case, but only a calculation can tell.
I’m here to make money trading my account. Either the solar max is a fundamental influence on the financial markets or it isn’t. I see a wealth of evidence that it is, and believe anyone who doesn’t pay attention to the correlations is down a very important tool with which to navigate the markets. The practical application with a near term view isn’t easy, but I see this as a step by step process. If you are not convinced I am fine with that – I am not selling anything. Time will reveal.
Sorry, but there is no “wealth of evidence” when there are only 20 observed solar cycles.
It would be like studying only one or two years of lunar cycles, which would be 20 cycles, and then talk about a wealth of evidence for whatever patterns we might find in those two years.
In proper research, even 200 observed cycles is still generally considered weak evidence, unless a very strong correlation is found (like 0.80 or more).
How strong is the correlation we see in your long term inflation chart?
One quick way to find out is by calculating the average distance of all the inflation peaks to the peaks in solar cycle. The average distance should be smaller than 2.75 years if there is any positive correlation with the 11 year solar cycle. In this case we can see too many peaks are two or more years away from solar peaks, so the correlation is very weak.
Click to access Scafetta_JStides.pdf
When you get some time, take a look at these pieces from Ian Wilson and Nicola Scafetta.
Both seek to prove that planetary tidal forces influence the sunspot cycle. Scaffeta predicts a new “Dalton style” minimum ahead, and accounts for the variances in the length of the Schwabe cycle.
Thanks TDL – see the charts in my post here: https://solarcycles.net/2012/02/23/solar-cycles-and-astro-trading/
Venus, Jupiter and Earth – Saturn was not included, but I guess the key thing is link between planets and stocks through the sun
I see you are ahead of me John…lol.
I though the paper by Scafetta was interesting though, especially this bit:
“We believe that our result is already quite important because
it elegantly rebuts a fundamental critique proposed by Smythe
and Eddy (1977). These authors argued that Jupiter and Saturn’s
planetary tides during the solar Maunder minimum present
patterns indistinguishable from those observed in the modern
era. Thus, it was concluded that planetary tides could not
inﬂuence solar dynamics. Essentially, planetary tidal patterns
were found uncorrelated with the solar records, which present
complex patterns such as the Oort, Wolf, Sporer, Maunder and ¨
Dalton grand solar minima plus the quasi-millennial cycle, etc.
The above critique in the early 1980s deﬁnitely convinced
most solar scientists to abandon the planetary theory of solar
variation ﬁrst proposed in the 19th century. In fact, while good
correlation patterns stimulate researchers to look for a possible
explanation and their physical mechanisms, a lack of correlation
can be easily interpreted as if no physical link truly exists!
On the contrary, we have found excellent correlation patterns
and demonstrated that the observed decadal, multi-decadal, secular
and millennial solar cycles can easily emerge as interference
patterns between the two major tidal cycles induced by Jupiter
and Saturn plus an internal solar dynamo cycle, which in our
model is indicated by the 10.87 year central Schwabe harmonic.
The 10.87-year period is about the average between the 9.93-year
and 11.86-year tidal periods, suggesting that a central internal solar
dynamo cycle may be approximately synchronized to and resonating with the average beat tidal period of Jupiter and Saturn.”
Being able to predict the length of the Schwabe cycle by the planetary alignments of Jupiter and Saturn could have practical benefits from a market timing perspective could it not.
It might also serve to explain the oscillations between equity/ commodity secular bull markets, and crucially, whether the pattern observed in recent history (ie. the last 100 yrs) may change with events like the Maunder minimum.
Ps Thanks for the sterling work you share on this blog.
Two further points:
1. Of course the lack of stock market data will be a issue if trying to track correlations over longer periods.
2. During these minimums (Dalton, Maunder etc.) would it not make sense for commodities to be of more relevance to humans? I mean energy/ food etc. would likely be far more important than “paper” assets if a large percentage of the world’s population is faced with a colder climate to navigate.
Just some thoughts!
Cheers TDL. Yes, all links together.
Commodities at the maunder and dalton minima, courtesy of Danny:
Prices were depressed for the maunder minimum, but the opposite during the dalton 1800-1820 (believe Danny has misplaced the red line which points towards 1850). So one of each.
Europe and the America’s were subject to devastating plagues at the time of the Maunder minimum, which might explain why commodities didn’t shine?
Hi John. Which line would you consider misplaced? The thick red line from Dalton minimum in the small chart?
Prices were depressed from ~1810-1845, which is well in line with 15 years after 1800-1820 (Dalton minimum).
So the Dalton minimum did create the depressed prices just like the Maunder minimum did. The 15 year lag time comes from the earth’s thermal inertia, heat stored in the oceans.
Hi TDL. Yes, population declines (from disease, war and famines) had a long term depressing effect on commodity prices in these periods. In some areas over 30% of the population died. That means there was much weaker demand for about everything, but this was interrupted by frequent spikes in food prices whenever famines and dearths occurred.
Maunder minimum 1650-1700 your red line points to the middle of this period on the commodities prices. Dalton 1800-1820 your red line points to 1850 on commodity prices. Prices were particuarly high 1800-1820, the opposite to around 1850.
Hi John. The thick red lines were only meant to mark the broad periods of maunder and dalton minimum, not particular years. The year by year relation is given in the lower right part of the chart. Wikipedia gives the Dalton minimum as 1790-1830, which is also very close to the sharply declining period in 11 and 22y ssn chart. If we add 15 year delay to see the effects of this Dalton period, then we get 1805-1845. The crb in 1845 was around 25, a 70 year low and more than 50% below the level at the start of the Dalton minimum. Even those who don’t buy the 15 year delay will find that in 1830 the crb was around 30, also 50% below the peaks around 1800. You are going to make the case that prices went up as a result of the Dalton minimum?
People who want to understand my chart will have to read the article.
Sunspots growing… perhaps my forecasts arent so bad after all…
SPX is now doing the fourth rally since 2009, and I really didnt think they were able to, considering the weakness around elsewhere
The three previous rallies were about 6month long and 300 points in SPX
-first one was longer, but can be counted from summer 2009 to New year
-sideways to summer
-rally to winter 2011
-sideways/down to fall
-rally to winter 2012
-sideways to fall 2012
-and now we have rallied ~6months and ~300 points since Nov
My 6month cycle suggested a top around April 12. A bearish wave would typically top earlier around March 12 (and Norway topped there). However we can also get extencded waves topping one month later, which gives around May 12.
So SPX is doing an extended wave in this cycle, and should soon find the cycle-low probably in June sometime (normally we should see a 3-4 week correction flushing down below 55ema somewhere. Near 5% below 35MA would be normal. Generally we should have reached the upside potential here at SPX 1630, and my downside targets are around 1500 +/-
I like your analysis…should be an interesting few weeks. I’m not buying the many media stories about how May 2013 is different, that’s for sure!
I have done some shorter timeframe cycle analysis to complement my 17.6 Year Stock Market Cycle (which has a low point in 2013).
The second chart in the link shows the third harmonic of the 17.6 WEEK cycle matching interim tops in the Dow almost exactly during the 2009 uptrend. It points to a top now and a low by 13/9/2013 (one day before St leger Day 🙂 )
Kerry, how does this 17.6 week play out prior to 2009? And do you have a ‘reasoning’ for this cycle existing?
No reasoning as yet John, still working on trying to identify the reason behind the 17.6 year cycle. The perigean cycle is 8.8 years (2*8.8 = 17.6) and TDL put me onto this:
but I am still trying to connect the dots.
Great gold trade btw, keep up the good work. Hopefully H2 2013 will be very interesting.
PS – The 17.6 week cycle was inverted in 2008/9 i.e. lows, following a 2007 high but the timing was a couple of weeks off. I will provide an updated post going back to 2000 when I have time.
Great stuff as usual. Danny your thoughtful observations are also powerful. Dealing with long term cycles I agree are hardly practical in terms of trading, but I believe that the hope in theory is to determine where we are in the longer term cycles. This would give more power and precision to shorter term indicators we rely on from a technical basis. I’m not throwing in the towel just yet.
The Yellow Boxes in each instance above correlate with recessions. The Saturn Neptune angles I notice suggest consistently historical panics that tend to land within a very tight orb of almost exact angles (0 degrees, 90 degrees, 180 degrees etc.) while recessionary economic down turns seem to manifest on odd degrees of separation. This information is pretty useless when trying to figure out which stock to buy, or when to sell. But it could be useful in determining which asset class to be exposed to, generally speaking. Weightings and timing need other indicators. I think your EARL indicator is showing valuable sensitivity for example Danny. In addition to the Saturn-Neptune cycle, which is 35.9 years in length, there are 6 other cycles not counting the Moon Phases and Solar Cycle. A composite of these other cycles suggests that John is onto something here despite an absence of overwhelming evidence supporting the facts. However, that said, you too suggest a valid point about deflationary effects not being fully absorbed by the markets just yet because the composite graph heads down later this year and doesn’t bottom until 2015 and then a second bottom in 2016. So, I agree with your observation as well that the inflation run up is well ahead and perhaps markets will experience drag related to damage taken on by the financial crisis, fiscal austerity and debt overhang around the globe.
What I do see in the first chart is that the Yellow Boxes mark recessions. This is interesting. I always believed that the market ‘discounted’ the economy and that policy lagged because it is reactive to economic needs not proactive. My hope is that the potential of the Solar Cycle may enable us to steer with more confidence. At the very least we may be able to decide on the right direction. But again, I think the magnitude of the moves are still in question. The other question is how much of a break do we make from what the historical data suggests regardless of the cycle we are looking at. All in all I feel very supportive of the work here.
Interesting article on historical efforts to reverse government spending and to implement austerity plans that are reforming Britain…
Appreciate your thoughts HVA.
Reblogged this on thedarklordblog and commented:
More sterling work from John Hampson…keep it up John!
Considering whether we are past cycle high or not one can look at solar polarity chart (http://www.solen.info/solar/polarfields/polar.html). This could be one argument for high being still in the future.
I have a problem with Bianco-Research commodity chart shown by Danny. You may call me radical from perspective of current economy but for the most of history silver was money and not commodity. Bianco index is including silver and gold. What is much more interesting is to follow the prices of consumable commodities like energy or food in monetary unit of the time. So for example looking at medieval times it would be wood for silver and so on.
Just to demonstrate my point I like to direct your attention to some historical data.
Looking at chart zoom and years between 1790 – 1820 you could have impression that commodities were getting cheaper. It was not really the reality if you look at grain prices in England.
1) 1799-1800 were so called famines years (price on the level of 160% of 25 annual average)
2) 1808-12 so called bad years (~150%)
3) 1816 was year without the summer (~130%)
Same story if you look at prices elsewhere for example France
1790 – 106 (grams of silver per 100kg grain)
1800 – 116
1810 – 144
source – Abel, Agrarkrisen und Agrarkonjunktur, appendix
Many scholars are suggesting that high price of commodities was one of the reasons for French revolution. There is substantial disconnect between what those numbers are saying and what Bianco-Research index is suggesting. Therefor I find it hard to believe in any concussions based on it.
Hi Pawel. I think they will eventually discard sunspot count in favor of polarity switch to define the solar maximum. That would remove a lot of ambiguity in declaring the solar cycle peak.
The problem with changes in currency/monetary unit over time is a real one and is always present, not just in historic charts and reconstructed charts, but even with more recent long term charts. If we look at a long term gold price chart in yen or euro we might not even recognize it. It looks totally different and the major peaks and bottoms can be years away from where they are in US$ gold price chart.
The recent rally in the nikkei index looks spectacular on the chart, but when we remove the effect of the declining yen it suddenly looks very similar to what other markets have done this year.
The chosen denominator (currency/gold/silver….) will always affect charts, because in reality a chart is always showing us how the prices of two different things have moved relative to each other. Price is always relative.
Abstracting from monetary units – my issue with Bianco index is that it suggests that commodities were getting cheaper just before French revolution.
So why all those crazy peoples risk their lives to get some of those supposedly cheap commodities. Now 200 years later Mr. Bianco is drawing index and claims commodities were all cheap at that time. It seems like a disconnect – somebody is wrong here. I don’t know for sure I can only read about those times but seems people life was quite cheap then and it was expensive to buy bread.
Anyway it would be nice to know a bit more about this index so that I can check what prices were actually falling in those years leading to and during revolution.
Hi Pawel. You could ask Bianco research how they have reconstructed this chart. As far as I know it is an average of several major countries, not just for France. The inflation peak that comes before the start of French revolution was connected to the American revolution. The hyperinflation in France peaked around 1795, as you can find here: http://wiki.mises.org/wiki/Inflation_during_the_French_Revolution
As far as I can see the Bianco research chart lines up well with that, the reconstructed crb doubled between 1789 and 1798.
I like the polarity chart…thanks for that.
Regarding commodities, and especially food and grains, a colder climate in the more northerly regions must surely make it difficult to produce high yields, thus driving up prices (barring any population catastrophies.) This does makes sense IMVHO.
One question. Where can I find frequently updated sentiment data. I look at ticker_sense but I would like to know what you are finding useful.
Sentimentrader; Daily Sentiment Indices (I am kindly provided these by someone); Investors Intelligence
However, lately I have been having doubts as to how useful the sentiment data is.
Interesting. Do you mean they lately become less useful or do you mean you lately doubt about they usefulness altogether ?
Just side question – As you have an access to those sentiment measures did you ever check their spectrum analysis ? Is there 11 years periodicity is there 2 weeks periodicity as suggested by Danny and his green/red zones ?
No I haven’t – good idea Pawel – have made a note to do that.
My issue is overbullish or overbearish can sustain for a long time. Nat Gas was overbearish for more than a year before turning. The Nikkei and Yen have been at extremes now for several months. It can just mean powerful bull or bear rally.
please check gorbanev.com