A great silver infographic..

A visual description showing how much silver has been mined in human history, 

http://www.silverseek.com/article/silver-visualized-bullion-bars-10046  

Why I am angry at having a 20% return in one day

This morning I was listening to NPR when I heard that Warren Buffett and a private equity group had just bought Heinz Corporation, the ketchup company, for cash, and that the Board of Directors had agreed to sell out for a price of 20% over previous closing.

My first reaction was that me and other long-term shareholders (I have held for over 10 years) were being taken for a ride.  The more I have read today, the more I am certain that other factors are in play here.  I don't know what these are, but this clearly could not have been for sound business reasons.  The board needs to be sent to the farm for such a dismal premium.

The reason I bought HNZ is because the business has so many safeguards against inflation or recession.  It is a staple product found in restaurants and refrigerators.  McDonald's condiment packets, Ore-Ida potatoes, Smart Ones.. it is a staple business that is everywhere without being recognized overtly... but when it is absent or replaced with cheaper brands.. people notice.

That is what makes this a brilliant brand.  How do you be everywhere, implicitly recognized as quality, but not visible enough to trip anyone's radar?  Freakin genius.

In terms of inflationary pressures HNZ is susceptible to a wide range of commodity price pressures, including corn yields and futures trading for the corn syrup in their mainline products.  But while shoppers may readily watch the price of gas at the pump like a hawk.. bleating when it goes up 10 cents a gallon (or ditto for a gallon of milk).. no one notices when a bottle of Heinz goes up in kind.

Did I mention this is a brilliant brand?

It has been able to raise its price over the years to keep margin over production costs steadily growing.  And its marketshare remains strong.

Partly due to activist investor Nelson Peltz, Heinz has also improved its operations aggressively and pressured the Board to increase its dividends as well.  Growing at double-digit rates these dividend payments helped buoy a rather attractive stock price that started clocking steady returns, year after year.

What makes me angry here is that based upon the past 4-5 years of activity, Heinz Corporation would have reached this magic 20% premium by 2014 or 2015, just a couple of years from now.  There has been nothing untoward in its business model and its internal growth would have kept the stock on track towards the $72 buyout price.

The Board is malfeasant.  No one sells out an entire company and betrays long-term shareholders for a price you would have hit in 8-12 quarters with business as usual.  It is unconscionable that they would have found a 20% premium to be in the best long-term interests of the shareholders.

I hope that class-action suits are filed and these Directors held accountable for their actions.  There has been nothing disclosed by the Company in any recent filings that would indicate Heinz has reached a peak valuation or is unlikely to continue growth in the next several years.  Something is afoul in Pittsburgh, and it ain't the Monongahela.

So that's it world.  I'm pissed for getting an extra 20% pop today.  So mad.. I may actually buy Hunt's now in protest.

The Mauna Loa Inflection: 400 ppm in 2014

Mauna Loa is the largest volcano by volume on Earth, one of the five volcanoes that form the Big Island of Hawaii.  And near its summit at over 11,000 feet above sea level lies a complex of instruments that collect atmospheric data, including levels of carbon dioxide.

This has been done more or less every day since 1958, and is considered the scientific standard for the globe since (1) being in the middle of the Pacific Ocean it is far away from all major land-based sources of pollution, natural and man-made, and (2) at two miles into the atmosphere is not affected by lower-atmospheric circulation.  It is literally, the temperature of the planet as far as these readings are concerned.  

Fact #1:  Physics dictates that carbon dioxide be a greenhouse gas.  

Carbon dioxide is an output gas of all breathing animals, decaying matter, and combustion of fossil fuels.  Carbon dioxide is a highly stable molecular arrangement of one carbon atom binding with two oxygen atoms.  It is very difficult to split apart because the nuclear bonds are so strong.  

Whenever you have strong bonds in a molecule you can add more heat than molecules with less stable bonds before it splits apart.  The 'heat capacity' of carbon dioxide means that when it absorbs energy from the sun, it 'holds' on to a part of that energy moreso than oxygen and water vapor (graph):  

The graph above shows relative heat capacity of major gas components of our atmosphere.  Carbon dioxide is at all temperature levels capable of storing more heat than its nearest competitor, water. 

Fact #2:  The amount of carbon dioxide gas is rising

It is an established scientific fact that the amount of carbon dioxide gas in our atmosphere has been rising:

In the above graph the red line is the monthly readings of carbon dioxide since 1958 (horizontal axis), whereas the black line is the trend line.  The vertical axis corresponds to the measured level of the gas, expressed in parts per million. 

This upward trend in readings looks steady but this rise when viewed in context with longer time frames is striking.  The below graph shows estimates of carbon dioxide readings as measured in Antarctic ice cores, going back a thousand years:

Prior sustained warming periods that are related to increased levels that took centuries and millennia... not years or decades.  

Building a Simple Forecast Model

I downloaded the monthly data from the National Oceanic and Atmospheric Association (NOAA).  

Next I input and conditioned the data for a Time Series model in IBM-SPSS Modeler, which located a Winter's Multiplicative best-fit:

This model excluded the first 12 months, as well as the most recent 48 months for holdout verification.  Additionally the model was instructed to iterate forward an additional 24 months for look-ahead forecasting.

The figure above shows the model results against the actual historic data.  Notice at the far right the 'blue' segment;  this is the three-year look-ahead forecast.

For clarity, the below shows the same series as above, but restricted to all time periods after the year 2000, this time looking ahead a full 48 months.

In this close-up view above, the 'blue' shaded area represents the forecast range.  Notice that it gets wider the further out in time you look;  this is a natural effect of the simple time series method being applied.  But overall the future pattern looks to fit the historic series with decent accuracy.  

The level of 400 parts per million (ppm) for carbon dioxide is a dangerously high level.  In terms of planetary chemistry it signifies that the heat retention capacity of the atmosphere has now reached very high levels.  When you have high heat capacity in the atmosphere roiling about, increased volatility in localized weather patterns is far more likely. 

The level of 400 ppm has never been seen at the Mauna Loa Observatory (but it has started to sporadically be measured in other parts of the globe).  

According to the forecast model above there is a 1 in 20 chance it will cross that boundary in Q2 of 2013.

NOAA's Earth Systems Research Laboratory (ESRL) in Boulder, CO puts the time of this threshold several years later, however:

“We will likely see global average CO2 concentrations reach 400 ppm about 2016.” -- Pieter Tans.

One problem with any statistical forecast is uncertainty, of unexplained variance.  This means that despite all the best techniques, there is always going to be some margin of error.  

But the margin of error can also be analyzed for patterns and trends.

This is where it starts to get interesting.

We now have two series of data for each month, the historic reading and the forecast value.  By taking the historic reading and subtracting the forecast value we get a residual value:

In the above I highlighted a few months data (2012-Q4) to show (1) historic data, (2) model forecast, and (3) the residual value.  

The values for residuals are all positive.  This means that the model underestimated the actual levels that were recorded at Mauna Loa.  

What is important to identify is whether or not the model is underestimating the actual data for long periods of time, or if this is a temporary effect.

To look at this scenario I plot all the residuals for every month.  Keep in mind that residuals are a measure of my forecast error;  positive values mean that the actual readings were higher than the forecast and negative values mean my model over-estimated the actuals.

This should not look like much to the casual eye.  In order to make sense of this we want to draw a line that represents the 'average' of all the points, as below:

What I see in this graph is interesting.

If my model were perfectly in balance in terms of forecast errors, then this line should be perfectly straight.  A straight, level line would mean that my model both over and underestimates actual readings in balance to each other, and that the rise in readings is happening more or less in steady fashion.  

But looking closer at the graph there is a discrete inflection point:

The arrow is pointing to the time period of late 1999, where the residual line (based upon decades of data) inflects or bends upwards.  This means that after 1999, the forecast built with decades worth of data appears to be underperforming the actual readings taken at Mauna Loa.

When this line bends up it may likely mean that the increase in carbon dioxide levels is accelerating beyond forecast parameters.  

If this is correct then it means we could hit 400 ppm earlier than 2016.  The only way 2016 could be the correct date would be if that residual line above is straight, and that the future rise in levels will continue more or less as it has for decades.  And the chart above shows that since 2000 the recorded levels have been higher than forecast.

I re-ran the model just for 1999 to end of 2012, to see what a shorter-term forecast would show, this time holding out the earliest and latest 12 months data for as holdouts:

In the above table are a few rows highlighted showing that 400 ppm is likely to be reached with good confidence around Q2 of 2014.  

This would be 1-2 years ahead of official forecasts for 400 ppm, and underscores the need to convert our energy and transportation systems away from fossil fuels and towards more renewable sources.  

This windfarm on the south tip of Hawaii won't power the island, but it is the direction we need to move towards at a pace rivaling the Manhattan Project, or the Apollo Program.  

There is no debate in the scientific community about climate change.  The evidence is more sound than the science used for most medical procedures today.  The only serious debate should be how to construct a new world that is less dependent upon systems that generate excess carbon dioxide, and rejuvenate the potential of ecosystems to absorb excess greenhouse gas already in circulation.