Political Calculations
Unexpectedly Intriguing!
March 22, 2017

Since we first uncovered the importance of dividend futures for anticipating future stock prices, we've been keeping track of that data on a regular basis.

Doing that has been a valuable exercise, not just because of our primary use for the data, but also because it gives us the ability to recreate what the future expectations of investors were on any trading day in the past for which we have the data.

For the Chicago Board of Exchange's dividend futures contracts, that history extends back to when they were first rolled out back in early 2010. We thought it might be interesting to compare what the future for dividends in each quarter looked like on the first day that the dividend futures contract for a given future quarter went into effect (one year before the futures contracts for it expires on the third Friday of the month ending the quarter for which it applies) and the value of the amount of dividends paid out during that quarter on the last day before its dividend futures contract expired.

The results are presented in the following chart:

First Day Forecast and Last Day Values Recorded for CBOE Dividend Futures, 2011-Q1 through 2017-Q1, with Forecast Values Through Quarters Ending in 2018-Q1

The chart begins with a comparison of the results between the dividend futures data for 2011-Q1 that was first published in March 2010 and the value of the dividend futures contract indicating the amount of the S&P 500's dividends per share would be paid out by the time that the dividend futures contract for 2011-Q1 expired on the third Friday of the March 2011.

Aside from the CBOE's dividend futures contracts' first year and a half, where the final day's dividend values had risen well above their initially recorded values, we see that there has generally been pretty close agreement between the initially forecast value and the value recorded on most dividend futures contracts' final day.

There are however two exceptions that stand out. The first is the event we've describes as "The Great Dividend Raid of 2012", where if you want to find out more about how the fear of higher dividend tax rates that were set to take effect in 2013 caused dividend payouts (and stock prices) in 2012-Q4 to surge so much higher than their initially forecast values, be sure to follow the link!

The other exception applies to the quarter whose dividend futures contract just expired last Friday, 2017-Q1, which has coincided with much of what the financial media has called the "Trump Rally", where the better than initially forecast results appear to have been fueled by improved earnings.

Sometime in the near future, we'll use our historic data on future expectations to take a closer look at the so-called Trump Rally.

Data Sources

EODData. Implied Forward Dividends March (DVMR). [Online Database]. Accessed 21-Mar-2017.

EODData. Implied Forward Dividends June (DVJN). [Online Database]. Accessed 21-Mar-2017.

EODData. Implied Forward Dividends September (DVST). [Online Database]. Accessed 21-Mar-2017.

EODData. Implied Forward Dividends December (DVDE). [Online Database]. Accessed 21-Mar-2017.

Labels: , ,

March 21, 2017

How well can dividend forecasts for individual stocks predict their future prices?

That's a question that we took on earlier this year, where we resuscitated an old stock price prediction technique to see how well it might work to identify potential investment opportunities for five companies, whose stocks had been identified as strong candidates for dividend increases during 2017-Q1. The five stocks and their dividend increase forecasts were:

  • Comcast (NASDAQ: CMCSA) - 15% increase
  • Home Depot (NYSE: HD) - 14% increase
  • Cisco Systems (NASDAQ: CSCO) - 15% increase
  • TJX Companies (NYSE: TJX) - 16% increase
  • Vulcan Materials (NYSE: VMC) - 25% increase

We then took annual dividend data and the stock prices recorded for each of these firms on the first day of trading in each year going back over the last several years, and plotted a relationship between them. We then used that relationship to identify if any stock prices in early 2017 were either potentially overvalued or undervalued with respect to the trends we identified. And then we sat back until we had results to see how the stock prices for each changed after they declared their dividends during the quarter.

We now have the results for 4 of the 5 companies, where we're only awaiting TJX's next dividend declarations to officially close the books on this experiment. The following chart shows the historical data we used to generate the trendlines for each stock, their forecast annual dividend for 2017, where their stock prices went on the day they declared their dividends and also where they were as of the close of trading on 20 March 2017.

Projected Trends for Stocks Forecast to Boost Dividends in 2017 Forward Annual Dividends Per Share vs January Price Per Share, Update 2017-03-20

In our original forecast, we had only identified two stocks that offered any sort of investment potential based on the value of their stock prices with respect to their trendlines: Home Depot (NYSE: HD), which looked to be undervalued, and Vulcan Materials (NYSE: VMC), which looked to be overvalued.

In the nearly two months since that analysis, we see that both assessments were correct. In VMC's case, we see that as VMC's dividend came in as expected, its stock price has moved downward to converge with its trendline. In Home Depot's case however, we see that company's dividend hike was far in excess of what had been forecast for it as the company reported a blowout quarter, where its stock price has risen, but not as strongly as its trendline would have suggested it might. And in truth, it has been diverging away from its historic trendline even as it has risen, which perhaps suggests that older trend may no longer be in effect and needs to be redrawn.

There was one real surprise for the quarter where these five stocks were concerned. Bucking the best-laid plans of professional dividend forecasters, Comcast (NASDAQ: CMCSA) pulled a fast one and split its stock on a 2:1 ratio, halving its dividend and very nearly also its stock price! The fun part of that change is that the stock price appears to have continued to track along its trendline, although it is now on the high side of it, suggesting that the stock is somewhat overvalued. Meanwhile, the stock price for Cisco Systems (NASDAQ: CSCO) behaved almost exactly as expected, keeping comparatively close to both its forecast and its historic trendline.

After TJX has declared its next dividend, which we think will be in April, we'll update the original version of this post on our site.

Labels: , ,

March 20, 2017

The third week of March 2017 was an up week for the S&P 500, where the index closed at 2,378.25 on Friday, 17 March 2017, just 5.65 points (0.24%) above where it closed at the end of the prior week of trading.

In the middle of the week that was Week 3 of March 2017, the Fed followed through and did exactly what it had been strongly signalling that it would do over the last several weeks, and boosted short term interest rates by a quarter percent, putting the Federal Funds Rate into a range between 0.75% and 1.00%.

From our perspective, both events were expected, where stock prices continued to fall within the range we first predicted two weeks ago (but didn't explain until last week).

Alternative Futures - S&P 500 - 2017Q14 - Standard Model with Connected Dots Between 2017-03-03 and 2017-03-21 - Snapshot on 17 March 2017

That doesn't mean however that we weren't surprised. For the Fed to have drawn such a strong focus where it would be using its influence to affect the expectations that investors have for the future, we should have seen a stronger upward movement in U.S. stock prices on Wednesday, 15 March 2017 than what the market saw.

That's because nearly all the Fed's officials have been doing everything they can to set the expectation that they will be hiking U.S. interest rates again in the very near future, which would put the timing of their next hike sometime in 2017-Q2, and which would most likely be announced at the end of the FOMC's June 2017 meeting.

Given where stock prices have been, since the expectations for the change in the year over year growth rates for S&P 500 dividends would place the S&P 500 higher than it currently is, for the Fed to have been successful in setting that focus would have coincided with a much larger jump in stock prices than what was recorded on Wednesday, 15 March 2017. At the very least, it should have moved to the middle of our echo-effect adjusted forecast range (indicated by the red-lined box on the chart above).

We would have expected that result because we have observed exactly that kind of behavior in stock prices following previous FOMC meetings where the Fed has directed the attention of investors to focus upon specific points of time in the future. So much so that we've used the timing of the FOMC's announcements to check the calibration of our dividend futures-based forecasting model.

So what gives now in Week 3 of March 2017? Why would stock prices appear to be behaving differently now than they have on the occasion of the Fed's previous announcements? Why would stock prices stay so much lower than what we would expect from even our echo-effect adjusted forecast?

We think the key to understanding what's going on with stock prices came out during Janet Yellen's post-FOMC announcement press conference, where Bloomberg's Kathryn Hays asked a pointed question:

KATHLEEN HAYS. Chair Yellen, Kathleen Hays. Oh excuse me, Kathleen Hays from Bloomberg. I'm going to try to take the opposite side of this because, on this question about market expectations and how the markets got things wrong, and then how you say the Fed suddenly clarified what it already said. But, for example, if the--if you look at the Atlanta Fed's latest GDP tracker for the first quarter, it's down to 0.9 percent. We had a retail sales report that was mixed, granted the, you know, upper divisions of previous months make it look better, but the consumer does not appear to be roaring in the first quarter, kind of underscoring the wait-and-see attitude you just mentioned. If you look at measures of labor compensation, you note in this statement that they're not moving up. And, in fact, they are--and if you look at average--there are so many things you can look at. And you, yourself, have said in the past that the fact that that is happening is perhaps an indication there's still slack in the labor market. I guess my question is this, in another sense, what happened between December and March? GDP is tracking very low. Measures of labor to compensation are not threatening to boost inflation any time fast. The consumer is not picking up very much. Fiscal policy--we don't know what's going to happen with Donald Trump. And, yet, you have to raise rates now. So what is the, what is the motivation here? The economy is so far from your forecast, in terms of GDP, why does the Fed have to move now? What is this signal, then, about the rest of the year?

CHAIR YELLEN. So, GDP is a pretty noisy indicator. If one averages through several quarters, I would describe our economy as one that has been growing around 2 percent per year. And, as you can see from our projections, we, that's something we expect to continue over the next couple of years. Now that pace of growth has been consistent with a pace of job creation that is more rapid than what is sustainable if labor force participation begins to move down in line with what we see as its longer run trend with an aging population. Now, unemployment hasn't moved that much, in part because people have been drawn into the labor force. Labor force participation, as I mentioned in my remarks, has been about flat over the last 3 years. So, in that sense, the economy has shown, over the last several years, that it may have had more room to run than some people might have estimated, and that's been good. It’s meant we've had a great deal of job creation over these years. And there could be, there could be room left for that to play out further. In fact, look, policy remains accommodative. We expect further improvement in the labor market. We expect the unemployment rate to move down further, and to stay down for the next several years. So, we do expect that the path of policy we think is appropriate is one that is going to lead to some further strengthening in the labor market.

KATHLEEN HAYS. Just quickly then, I just want to underscore. I want to ask you, so following on that, you expect it to move. What if it doesn't? What if GDP doesn't pick up? What if you don't see wage measures rising? What if you don't, what if the core PCE gets stuck at 1.7 percent, would you, is it your view, perhaps, that if there's a risk right now in the median forecast for dots, that it's fewer hikes this year rather than the consensus or more?

CHAIR YELLEN. Well, look, our policy is not set in stone. It is data dependent and we're, we’re not locked into any particular policy path. Our, you know, as you said, the data have not notably strengthened. I, there's noise always in the data from quarter to quarter. But we haven't changed our view of the outlook. We think we're on the same path; not, we haven't boosted the outlook projected faster growth. We think we're moving along the same course we've been on, but it is one that involves gradual tightening in the labor market. I would describe some measures of wage growth as having moved up some. Some measures haven't moved up, but there's some evidence that wage growth is gradually moving up, which is also suggestive of a strengthening labor market. And we expect policy to remain accommodative now for some time. So we're, we’re talking about a gradual path of removing policy accommodation as the economy makes progress, moving toward neutral. But we're continuing to provide accommodation to the economy that's allowing it to grow at an above-trend pace that's consistent with further improvement in the labor market.

We're kind of in a unique position in that we recognized, very early in 2017-Q1, that the Fed could get away with hiking short term interest rates as they did during Week 3 of March 2017, with almost no negative effect from shifting the forward-looking focus of investors from 2017-Q2 (where it had been focused) to the nearer term future of 2017-Q1 because the expectations for changes in the year over year growth rate of dividends per share in 2017-Q1 have been nearly identical with the expectations for 2017-Q2 for months. That similarity between those sets of expectations has meant very little impact to stock prices for investors shifting their focus back and forth between the two quarters.

But that's also occurred as the lagging effects from the 2014-2016 economic slowdown have become more pronounced in the economic data, as described by Kathryn Hays in her question to Fed Chair Janet Yellen. That deterioration in the economic data is increasingly leading investors to set their attention toward 2017-Q3, which given where stock prices are today, would tend to weight them down, where investors are betting that the negative economic data will cause the Fed to back off from hiking short term interest rates again until that more distant future quarter.

The result then is a split in the forward-looking focus of investors, where they would appear to now be dividing their focus between 2017-Q2 and 2017-Q3 in setting today's stock prices, with the outcome of stock prices that have been mostly moving sideways to slightly upward.

In this environment however, otherwise minor news about that economic data that would ordinarily not be of much concern to investors can have an outsized impact. Good numbers on the economy will lead investors to favor 2017-Q2 for the timing of the Fed's next rate hike, with stock prices rising as a result. Bad numbers will tend to direct investors to focus on 2017-Q3, sending stock prices lower as they bet the Fed will delay its next rate hike.

That dynamic is setting up just as the S&P 500 is reaching a near record period of relatively low volatility, in which it hasn't seen a decline of at least 1% during the past 108 trading days. We think that's going to change in a very noticeable way in the very near future, especially as the underlying trajectories for the basic path of future stock prices is also about to change from generally upward to generally downward (more on that next week!) Until then, get ready to hang onto your hats!

The other news that stood out to us during the Week that Was Week 3 of March 2017 is linked below!

Monday, 13 March 2017
Tuesday, 14 March 2017
Wednesday, 15 March 2017
Thursday, 16 March 2017
Friday, 17 March 2017

Elsewhere, Barry Ritholtz recaps the week's positives and negatives for the U.S. economy and markets.

Labels: ,

March 17, 2017

One of the more surprisingly popular posts that we've featured this year was "A Centrist's Guide to Media Bias and Usefulness", in which we featured a chart that ranks general news reporting web sites by the information quality of their reporting and a more accurate indication of their relative position on today's political spectrum than an earlier version that had gone viral.

So we were intrigued when we found that RealClearScience's Ross Pomeroy and Tom Hartsfield had joined with the American Council on Science and Health's Alex Berezow (formerly of RealClearScience) to create a similar chart for science news web sites.

Although here, they adapted the measure of information quality to focus more on how strongly evidence-based the stories covered by the science news sites they evaluated were, which they then assessed against a non-political spectrum that assesses how accessible or compelling each sites' stories would be to a general audience. Here's the chart they developed:

ACSH-RCS Infographic: Evidence-Based Science Reporting vs General Readability, 2017

Overall, their rankings are pretty solid, though we might tweak the placement of some of the sources (we would rank Wired slightly higher on the "compelling" measure for example, although that's attributable to the kinds of science news stories that we find interesting.

As for what sites we'd like to see added to the rankings, we would put Phys.org on the list, although it has considerable overlap with LiveScience as a news aggregator, it covers more fields of science. Its downside is that a lot of the stories it picks up read like press releases, so for a lot of stories, it would only fall in the "sometimes compelling" categorization.

We'd also look to add more maths into the mix, which though it falls outside the focus on evidence-based science, where mathematics news is concerned, there are considerably fewer sites that generate anything close to what might make for compelling reading by a general audience. And really, we'd recommend just three for that kind of consumption: Quanta and +Plus, for written word reporting, and BBC Radio 4's More or Less, which is presented in audio format.

Previously on Political Calculations


March 16, 2017

Richard Feynman, offering amazing insight, back in 1981:

HT: Luboš Motl.

Sadly, not much has changed for social sciences in the intervening three and a half decades, where if the series of fresh junk science examples that we compiled during the last six months of 2016 is any indication, the same problems that Feynman noted back in 1981 are still pervasive in these fields - and the reason why is still the same: it takes much less effort to generate junk science than the care and attention it takes to generate the real thing!


March 15, 2017

It has been just over a year since we last featured an update to our chart that correlates the change in atmospheric carbon dioxide with the relative health of the Earth's economy.

We haven't been holding back because we didn't want to share what we were seeing - we were holding back because large El Niño-related wildfires in Indonesia in 2015 flooded the Earth's atmosphere with extra carbon dioxide, which meant that we couldn't tell the difference between the additional contribution of CO2 from wildfires and that contributed by productive human activities in telling how much that each was influencing the atmospheric CO2 data.

But we're now coming up on when that extra CO2 will have fully dissipated from the atmosphere, leaving it clean enough to more clearly communicate how the global economy is faring. Speaking of which, here is what we see today when we measure the trailing twelve month average of year-over-year change in the parts per million of atmospheric carbon dioxide as measured at the Mauna Loa Observatory.

Trailing Twelve Month Average of the Year Over Year Change in the Parts Per Million of Atmospheric Carbon Dioxide, January 1960 through March 2017

In this chart, we've begun to see the trailing year average in the rate at which the amount of carbon dioxide in the Earth's air is changing begin to decline, which will become more dramatic over the next several months once we get past the one-year anniversary of when the contribution of 2015's Indonesia wildfires to atmospheric CO2 measured at Mauna Loa peaked in May 2016.

The next chart looks at the raw year-over-year change in atmospheric CO2, where we can confirm that those levels have dropped to levels that are consistent with the upper end of the range where they've been through much of the current century. [Note: The spike in the recorded data for January 2017 appears to be something of an anomaly.]

Year Over Year Change in the Parts Per Million of Atmospheric Carbon Dioxide, January 1960 through March 2017

We think that is in part because the economies of both China and the U.S. have begun growing more strongly than they were a year ago, where China's economy in particular appears to be growing much faster.

That's significant because China's economy is, by a wide margin, the world's largest producer of emissions of carbon dioxide. If China's economy is truly growing, it shows up in the Earth's atmosphere in a big way.

Data Source

National Oceanographic and Atmospheric Administration. Earth System Research Laboratory. Mauna Loa Observatory CO2 Data. [File Transfer Protocol Text File]. Updated 6 March 2017. Accessed 12 March 2017.


March 14, 2017

The U.S. Census Bureau's data for goods and services exported to the U.S. from China and from the U.S. to China in January 2017 suggests that both economies are growing much more strongly than they were a year earlier.

At least, that's what is suggested by the sharp rise in the exchange-rate adjusted year-over-year growth rate of the value of goods and services traded between the two nations in the following chart!

Year Over Year Growth Rate of Exchange Rate Adjusted U.S.-China Trade in Goods and Services, January 1986 - January 2017

Before we continue, we should recognize that this data partially overlaps this year's weeklong Chinese New Year/Spring Festival holiday, which because the timing of the holiday is based on a lunar calendar, can greatly affect the year-over-year growth rate calculation for when the holiday falls entirely in the month of January or the month of February. This year, the Chinese holiday is straddling between the two months, where in 2016, it fell entirely within February.

Consequently, we anticipate that the data for the goods that China imports from the U.S. for February 2017 will show significantly higher growth. There is evidence that indicates that this is indeed happening, where China's preliminary trade data is showing such a remarkable surge in imports that they are outnumbering China's exports for the month:

China's imports surged 38.1 percent from a year earlier, the biggest increase since February 2012, official data showed on Wednesday, while exports unexpectedly fell 1.3 percent.

That left the country with a trade deficit of $9.15 billion for the month, the General Administration of Customs said.

Most analysts, however, attributed the rare trade gap to distortions caused by the long Lunar New Year celebrations, which began in late January this year but fell in February in 2016. Many businesses shut for a week or more and factory production and port operations can be significantly affected.

Looking at the U.S. side of the trade ledger, the preliminary data for February 2017 also suggests a year-over-year increase, which is expected to be followed by larger volumes of exports in March and April:

For January, the most recent month for which data is available, ports in the report handled what was cited as an “unusually high” 1.67 Twenty-Foot Equivalent Units (TEU), which was up 6.5 percent compared to December and was up 12.5 percent annually. This high volume level came on the heels of Asia-based factories shutting down in advance of the Lunar New Year.

The report pegged February at 1.61 million TEU for a 4.2 percent annual increase. But the larger volumes appear to be slated for March and April, which are estimated to hit 1.46 million TEU (a 10.6 percent annual gain) and 1.59 million TEU (a 10.1 percent annual gain), respectively. May is projected to be up 2.9 percent annually at 1.67 million TEU, and June is expected to see a 5.5 percent annual increase at 1.66 million TEU.

Some of the news reports on trade that we're seeing are also providing some estimates of how much the "border adjustment" tax being considered by the Trump administration would affect the trade balance, which we'll take a closer look at in upcoming weeks.

If you want to get there before us, you're more than welcome to do so - here's a tool that we built earlier this year to do some of the "tariff meets supply and demand" math, which you can pair with the appropriate trade source data.

Data Sources

Board of Governors of the Federal Reserve System. China / U.S. Foreign Exchange Rate. G.5 Foreign Exchange Rates. Accessed 12 March 2017.

U.S. Census Bureau. Trade in Goods with China. Accessed 12 March 2017.


March 13, 2017

The S&P 500 in Week 2 of March 2017 ended just ever-so-slightly below where it opened the week. After generally trending downward on the week, the S&P 500 recovered much of what the index had lost on Friday, 10 March 2017.

Our alternative futures chart updates the trajectory that the S&P 500's closing prices took in the second week of March 2017 against the backdrop of what our dividend futures-based standard model projected, based on a combination of historical stock prices and the expectations that investors have for the amount of dividends per share to be paid out through specific points of time in the future.

Alternative Futures - S&P 500 - 2017Q1 - Standard Model with Connected Dots Between 2017-03-01 and 2017-03-21 - Snapshot on 2017-03-10

That brings us to the one thing that really drew your eyes in this week's updated chart - that red box outlining where the S&P 500's closing stock prices on each day of the week fell. The one that we teased at the end of last week's edition!

Like the previous red-shaded box, the new red-lined box represents our prediction of the range into which the S&P 500 will close over the indicated period of time during the indicated period of time, which runs from 3 March 2017 through 21 March 2017. We've drawn both boxes on top of our standard model's projections in order to account for the effect of the echo of past volatility on our standard model's projections, which arises as a result of our model's use of historic stock prices as the base reference points from which we project future stock prices.

Unlike the previous red-shaded box shown on the chart, the starting point is not aligned with the projected trajectory that corresponds with the future quarter that we believe investors have fixed their forward-looking attention upon in making their current day stock prices, which we believe to be the current quarter of 2017-Q1 based on the Fed's recent concerted efforts to convince market participants that it will next hike short term interest rates in the U.S. before the end of the quarter.

In doing that, we're trying something new, which draws upon our observation that the trajectory of stock prices are very much like a kinematics problem from physics. In such a problem, the actual trajectory that a body in motion that is being acted upon by external forces might take is anchored by its starting position.

So what we've done is to shift the forecast range that corresponds to where our standard model projects for investors fixing their attention on 2017-Q1 (and also 2017-Q2, since that future quarter's expectations are nearly identical to those of 2017-Q1) downward, to adjust for the starting value of the period of time covered by our prediction. We should note that if investors should shift their forward-looking attention to a different point of time in the future with different expectations associated with it, such as 2017-Q3 or 2017-Q4, stock prices will move outside the predicted range indicated by the red-lined box.

One week into this particular forecasting experiment, it seems to be working, but we have another week and a half to go. Until then, here are the headlines that caught our attention during the second week of March 2017....

Monday, 6 March 2017
Tuesday, 7 March 2017
Wednesday, 8 March 2017
Thursday, 9 March 2017
Friday, 10 March 2017

For a bigger picture of the week that was, Barry Ritholtz lists the week's positives and negatives for the U.S. economy. If you're a fan of business and finance-oriented podcasts, you might consider checking out Bloomberg's Masters in Business series, which Barry hosts. Speaking of podcasts, we would also strongly recommend Russ Roberts' Econtalk (which recently posted its rankings of its most popular sessions for 2016, and also the BBC's brilliant 50 Things That Made the Modern Economy, which is presented by More or Less' Tim Harford and which features excellent production values while packing an amazing amount of information into its 9-minute-long episodes.

Labels: ,

March 10, 2017

The flights of birds have been inspiring human invention for thousands of years. And for thousands of years, people haven't had much success in mimicking the flapping motion of a bird's wings to produce the kind of thrust needed to power themselves into the air outside of very small drones.

But what if instead of trying to generate both the upward and forward thrust to fly, somebody tried to simply model the motion of a bird's flapping wings in flight to generate electrical power by more efficiently capturing energy from the wind?

Somebody in this case is a team of Tunisian engineers employed by Tyer Wind, who have developed a prototype wind turbine whose blades model the flapping motion of hummingbird's wings to generate electricity. Here's their promotional video that introduces the technology, which is made all the more impressive by its soundtrack.

Before anybody gets too excited, we should recognize that the only way that we would ever see a forest of pylons with flapping, power-generating wings mounted on top of them is if they can successfully generate more electricity both more efficiently and more economically than a more traditional wind turbine can produce.

Engineering's Tom Lombardo did some math to find out how much power Tyer Wind's prototype flapping wind turbine might produce:

Tyer hasn't yet released a wind curve - they're waiting for the results of field testing. Their data sheet indicates that the turbine's rated output is 1 kW at a wind speed of 10 m/s (22.3 mph). At that velocity, the wind carries just under 5 kW of power, so this turbine is roughly 20% efficient.

Keep in mind that the rated outputs of small wind turbines are specified for wind velocities that you'll almost never see in places where they're likely to be located. If your wind speed were a steady 10 m/s, all of your trees would be bent over at 45° angles.

With that in mind, let's estimate this turbine's output at a more realistic wind velocity: 5 m/s (11.2 mph) - a moderate breeze. Since wind power varies with the cube of wind speed, cutting the velocity in half results in an eightfold decrease in the available wind power, dropping the wind input to only 616 Watts. With 20% efficiency, this turbine would generate a paltry 123 Watts. It's probably even lower than that, since turbines are less efficient at slower velocities. But in the absence of a power curve, I'm giving them the benefit of the doubt and assuming 20% efficiency across the board. (For comparison purposes, I checked into a popular horizontal axis wind turbine with a 1 kW rating; its power output is similar at 5 m/s wind speeds.)

So the concept has some potential, but isn't the kind of clear home run that wind power enthusiasts might have hoped. It will be interesting to see the data from the prototype's testing.

Previously on Political Calculations


March 9, 2017

Previously, in looking at the total market cap of the new homes sold in the United States, we found that the market for new homes in the U.S. began to noticeably decelerate after September 2016.

Since market capitalization is the product of both new home sale prices and the quantity of new home sales, to find out why that outcome is happening, we first looked at the trailing twelve month average of new home sale prices. The following chart shows the most recent major trends in median new home sale prices in the period from July 2012 through January 2017.

Trends in Trailing Twelve Month Average of Median U.S. New Home Sale Prices, July 2012 through January 2017

In this chart, we see that after having risen at a relatively slow pace in the months from August through October 2016, they briefly dipped in November 2016, before rebounding to rise at a somewhat faster pace than they had in the months preceding November 2016.

But since the market cap was falling from October 2016 onward, that combination can only mean that the number of new homes being sold was falling more sharply than their sale prices were rising. And sure enough, we do see that after spiking in July 2016, the number of new home sales was rising slowly up to November 2016, after which they dropped sharply in December 2016.

Thousands of New Homes Sold in United States, July 2012 through January 2017, July 2012 through January 2017

Before we continue, we should note that the most recent three months of data shown in the charts above are still subject to revision, so the price and quantity data for new home sales during the months of November 2016, December 2016 and January 2016 may all change during the next several months.

Back to the topic at hand, one obvious candidate to consider that could potentially explain what we see in both the price and quantity data for new home sales is mortgage rates. The following chart shows how the average interest rates for conventional 30-year have changed from month to month from July 2012 through January 2017.

30-Year Conventional Fixed Mortgage Rates in U.S., July 2012 through January 2017

In December 2016, the average 30-year fixed interest rate mortgage in the U.S. rose by 0.43 points to 4.20% from the previous month, which just happened to be nearly three-quarters of percentage higher than where it had bottomed in July and August 2016.

But that is not the only factor behind the stalling of the new home market in the U.S.

Let's next update our chart showing the relationship between the trailing twelve month average of median new home sale prices and median household income, which we'll do for the period from December 2000 through January 2017.

U.S. Median New Home Sale Prices vs Median Household Income, December 2000 through January 2017

In this chart, we can confirm the generally rising growth trend for median new home sale prices, but we also see a slowing growth trend for median household income in the latter half of 2016. The following chart shows just median household income in both nominal (current dollar) terms and in real (constant January 2017 U.S. dollar) terms.

U.S. Median Household Income, Nominal and Real (Constant Jan-2017 U.S. Dollars), December 2000 through January 2017

Here, we see that both real and nominal median household income in the U.S., after having risen rapidly in 2015, began to decline through the first four months of 2016. Afterward, both nominal and, to a lesser extent, real median household income rose slowly through November 2016. Then, in December 2016, both nominal and real median household income declined.

January 2017 saw the nominal median household income rise, but after adjusting the household incomes in previous months for inflation, we find that in real terms, median household income fell again in January 2017.

The final chart we'll share today will look at the median household income data again, but this time, measured as the year over year growth rate for the nominal and inflation-adjusted data. Since we only have the monthly median household income data back to December 2000, we'll show the year over year growth rates from January 2001 through January 2017.

Year Over Year Change in the Rate of Growth of U.S. Median Household Income, Nominal and Real (Constant Jan-2017 U.S. Dollars), December 2001 through January 2017

Measured year over year, we see that the growth rates of both nominal and inflation-adjusted median household income in the U.S. began to sharply decline after March 2016. As of December 2016, the year over year growth rate of the inflation-adjusted median household income had fallen significantly into negative territory for the first time since early 2012.

So, the slowing market cap for new homes in the U.S. in the latter part of 2016 is a consequence of the combination of rising new home sale prices, substantially fewer new home sales (which is likely contributing to the rising prices), rising mortgage rates, and slowing or falling household income growth. With that kind of negative momentum, is it any wonder that the Fed has sharply reduced its nowcast for GDP growth estimates in 2017-Q1?

Data Sources

U.S. Census Bureau. New Residential Sales Historical Data. Houses Sold. [Excel Spreadsheet]. Accessed 8 March 2017.

U.S. Census Bureau. Median and Average Sales Prices of New Homes Sold in the United States. [Excel Spreadsheet]. Accessed 8 March 2017.

Freddie Mac. 30-Year Fixed Rate Mortgages Since 1971. [Online Database]. Accessed 8 March 2017.

Sentier Research. Household Income Trends. [PDF Document]. Accessed 2 March 2017. Note: All data is converted to be in terms of current (nominal) U.S. Dollars.

U.S. Department of Labor. Bureau of Labor Statistics. Consumer Price Index, All Urban Consumers - (CPI-U), U.S. City Average, All Items, 1982-84=100, Not Seasonally Adjusted. [Online Database]. Accessed 8 March 2017.

National Bureau of Economic Research. U.S. Business Cycle Expansions and Contractions. [Online Database]. Accessed 8 March 2017.


March 8, 2017

Nicholas Eberstadt recently raised an interesting question regarding the role of Medicaid, the United States' welfare program that provides free health insurance for Americans with low incomes, when he commented on how it may very well have directly contributed to the nation's rising problem with opioid addiction.

The opioid epidemic of pain pills and heroin that has been ravaging and shortening lives from coast to coast is a new plague for our new century. The terrifying novelty of this particular drug epidemic, of course, is that it has gone (so to speak) "mainstream" this time, effecting breakout from disadvantaged minority communities to Main Street White America. By 2013, according to a 2015 report by the Drug Enforcement Administration, more Americans died from drug overdoses (largely but not wholly opioid abuse) than from either traffic fatalities or guns. The dimensions of the opioid epidemic in the real America are still not fully appreciated within the bubble, where drug use tends to be more carefully limited and recreational. In Dreamland, his harrowing and magisterial account of modern America’s opioid explosion, the journalist Sam Quinones notes in passing that "in one three-month period" just a few years ago, according to the Ohio Department of Health, "fully 11 percent of all Ohioans were prescribed opiates." And of course many Americans self-medicate with licit or illicit painkillers without doctors' orders....

How did so many millions of un-working men, whose incomes are limited, manage en masse to afford a constant supply of pain medication? Oxycontin is not cheap. As Dreamland carefully explains, one main mechanism today has been the welfare state: more specifically, Medicaid, Uncle Sam's means-tested health-benefits program. Here is how it works (we are with Quinones in Portsmouth, Ohio):

[The Medicaid card] pays for medicine—whatever pills a doctor deems that the insured patient needs. Among those who receive Medicaid cards are people on state welfare or on a federal disability program known as SSI.... If you could get a prescription from a willing doctor—and Portsmouth had plenty of them—Medicaid health-insurance cards paid for that prescription every month. For a three-dollar Medicaid co-pay, therefore, addicts got pills priced at thousands of dollars, with the difference paid for by U.S. and state taxpayers. A user could turn around and sell those pills, obtained for that three-dollar co-pay, for as much as ten thousand dollars on the street.

In 21st-century America, "dependence on government" has thus come to take on an entirely new meaning.

It occurred to us that we have the ability to determine whether Medicaid is contributing to the nation's growing opioid epidemic by taking advantage of a natural experiment made possible by the Affordable Care Act. More popularly known as "Obamacare", the primary means by which the ACA has expanded health insurance coverage in the U.S. has been through the expansion of eligibility in the U.S.' Medicaid welfare program, where the threshold for eligibility was raised from 100% of the federal poverty limit to 138% of the federal poverty limit in states that agreed to expand their Medicaid programs.

But, not all states agreed to expand their Medicaid program when the Affordable Care Act went into effect on 1 January 2014. From that time through 2015, 28 states and the District of Columbia had chosen to participate in the Affordable Care Act's expansion of eligibility for Medicaid, while 22 others opted to not do so during those years.

That state-by-state division then provides for the basis of a natural experiment, where we can get a sense of how well correlated the expansion of enrollment in Medicaid is with the incidence of deaths from drug overdoses, which have been increasing as part of the U.S.' growing problem with opioid drug addiction.

The image below shows what we found for the age-adjusted death rate per 100,000 Americans from drug overdoses for the years from 2010 through 2015, which covers the years in which the individual U.S. states that chose to expand their Medicaid enrollment did so. We've ranked the states in order from the greatest increase in the rate of deaths from drug overdoses between 2013 and 2015, which for all but six states, marks the difference between "before" and "after" for the expansion of their Medicaid programs.

Age-Adjusted Deaths from Drug Overdoses Per 100,000 Population, 2010-2015

Overall, the biggest increase in the age-adjusted death rate from drug overdoses by state from 2013 to 2015 was an increase of 19.2 deaths per 100,000 (New Hampshire), and the biggest recorded decrease was 1.6 deaths per 100,000 (Oklahoma).

Looking just at the change from 2013 to 2015, the median change in the number of deaths from drug overdoses is 2.0 per 100,000 of the state's population. For the 28 states that acted to expand the eligibility of their state's Medicaid programs, 19 have seen the number of deaths from drug overdoses increase by 2.0 per 100,000 or higher per year, accounting for 68% of all Medicaid expansion states. The change in each of these states' age-adjusted death rates ranged from New Hampshire's biggest increase of 19.2 deaths per 100,000, to Nevada's decrease of 0.7 deaths per 100,000, where the median change was an increase of 2.7 drug overdose deaths per 100,000 population.

Median Change in Deaths from Drug Overdoses per 100,000 Population Since 2010, Medicaid Expansion States vs Non-Expansion States

For non-Medicaid expansion states, only 7 of 22 states saw their age-adjusted death rates from drug overdoses increase by 2.0 per 100,000 or more per year, representing 32% of the states whose increase in death rates exceeded the median for all states. In the non-Medicaid expansion states, the change in each of these states' age-adjusted death rates ranged from the biggest increase of 8.0 deaths per 100,000 population for Main, to a Oklahoma's decrease of 1.6 deaths per 100,000 population. The median change for the non-Medicaid expansion states was an increase of 1.5 deaths per 100,000 population.

We also looked at the change in the age-adjusted death rates for each state from 2010 through 2013. Here, we saw the change in death rates per 100,000 for each state range between a decline of 3.8 deaths per 100,000 (Florida) and a high of 6.9 deaths per 100,000 (Rhode Island), with a median increase of 1.6 deaths per 100,000 for all 50 states and the District of Columbia. The median change in age-adjusted deaths from drug overdoses in the states that would act to expand their Medicaid programs was an increase of 2.2 deaths per 100,000 from 2010 through 2013, while the median change for states that did not expand their Medicaid programs was 1.2 deaths per 100,000 during this period.

Those figures are significant because 2010 marked the year in which the use of opioid street drugs like heroin and synthetic versions like fentanyl and tramodol began to take off, along with deaths from drug overdoses as they were increasingly substituted for prescription medications like OxyContin by addicts, which came about because of the introduction of abuse-resistant forms of the opioid-based medication.

To put that increase in perspective, in 2010, heroin alone was responsible for 8% of drug overdose deaths in 2010. Five years later, heroin alone was responsible for 25% of all drug overdose deaths in the U.S. Deaths caused by overdoses of synthetic opioids have increased similarly.

That change would account for a portion of the increased drug overdose deaths across the entire nation from 2010 onward, but the state-by-state mortality data suggests those deaths accelerated after 2013, which means that something else changed after 2013 to drive the number of drug overdose deaths to increase at a rate so much faster than they had before the Affordable Care Act was implemented.

As we've shown, that acceleration took place predominantly in the states that had expanded their Medicaid programs, where in accordance with the policies and practices advocated from the top down by a federal government bureaucracy that saw prescribing painkillers as a less costly option when compared with more expensive medical treatments, many newly covered patients were prescribed the painkilling medications to which they would become addicted, at nearly no out-of-pocket costs to themselves.

Until the prescriptions ran out. That's when the newly addicted turned to the opioid substitutes available in the U.S.' black market, driving up the nation's opioid addiction rates to new highs in portions of the population that had not previously been at such elevated risk.

If access to health care through Medicaid really improved health care outcomes, these are outcomes that we would not observe. That the expanded access to Medicaid would appear to be making that problem worse indicates that the Medicaid welfare program is in dire need of major reform.


U.S. Centers for Disease Control and Prevention. National Vital Statistics System, Mortality. CDC WONDER. [Online Database]. Atlanta, GA: US Department of Health and Human Services, CDC; 2016. [Note: Deaths were classified using the International Classification of Diseases, Tenth Revision (ICD–10). Drug overdose deaths were identified using underlying cause-of-death codes X40–X44, X60–X64, X85, and Y10–Y14.]

Eberstadt, Nicholas N. Our Miserable 21st Century. Commentary. [Online Article]. 15 February 2017.

Kaiser Family Foundation. States Getting a Jump Shart on Health Reform's Medicaid Expansion. [Online Article]. 2 April 2012.

Center on Budget and Policy Priorities. Status of State Medicaid Expansion in 2015. [PNG Image]. 28 April 2015.

Bloom, Josh. Have Opioid Restrictions Made Things Better or Worse? American Council on Science and Health. [Online Article]. 3 November 2016.

Rettner, Rachel. US Drug Overdose Deaths Continue to Rise: Here Are the Numbers to Know. LiveScience. [Online Article]. 24 February 2017.

Labels: ,

March 7, 2017

About once a quarter, we calculate the market capitalization for the U.S. new home market, where we multiply the number of reported new home sales each month by their reported average sale prices. Once we have those figures, we calculate trailing 12 month average of these market cap figures to account for annual seasonality in the sales data, which we then visualize in the form of charts.

The results of that exercise can tell us quite a lot about the state of the new home market in the U.S., which for this edition, suggests that new home sales are stalling out once again, indicating that things are slowing down for the U.S. homebuilding industry. The following chart shows what we found in nominal terms.

Trailing Twelve Month Average New Home Sales Market Capitalization, Current U.S. Dollars, December 1975 - January 2017

The deceleration appears to have taken hold beginning in October 2016, when the rate of growth of the U.S. new home market cap suddenly downshifted to grow much more slowly, slowing in each month since through January 2017, as the market cap's trailing year average neared $17 billion.

The following chart shows the same data, but this time adjusted for inflation, so that all the historic market cap values for the U.S. new home market are shown in terms of constant January 2017 U.S. dollars.

Trailing Twelve Month Average New Home Sales Market Capitalization, Constant January 2017 U.S. Dollars, December 1975 - January 2017

In both charts, we see that this is the third time since the market for new homes bottomed in January 2011 that the growth of the new homes market cap has sputtered to a near stall after a period of relatively strong growth. We'll need to dig more into the data for 2016 to find out what the driving factors are behind the change since September 2016, which we'll follow up later this week.

Data Sources

U.S. Census Bureau. New Residential Sales Historical Data. Houses Sold. [Excel Spreadsheet]. Accessed 1 March 2017.

U.S. Census Bureau. New Residential Sales Historical Data. Median and Average Sale Price of Houses Sold. [Excel Spreadsheet]. Accessed 1 March 2017.

U.S. Department of Labor Bureau of Labor Statistics. Consumer Price Index, All Urban Consumers - (CPI-U), U.S. City Average, All Items, 1982-84=100[Online Application]. Accessed 1 March 2017.


March 6, 2017

For the S&P 500, the first week of March 2017 was about as eventful as it could be thanks to a strong attempt by Fed officials to shape future expectations ahead of their two-day meeting beginning on 14 March 2017. We would argue that through their concerted efforts, they succeeded in shifting investors to focus upon 2017-Q1 instead of 2017-Q2 to coincide with the timing of their next rate hike.

Alas, because the expectations for changes in the rates of dividend growth for the current quarter of 2017-Q1, which still has a few weeks left in it, and the upcoming quarter of 2017-Q2 are nearly identical, that shift in focus didn't do much to significantly affect the level of the S&P 500, which continued to track along near the bottom of the likely range that our dividend futures-based model has projected for the two quarters.

Alternative Futures - S&P 500 - 2017Q1 - Standard Model - Snapshot 2017-03-03

Last week was also interesting in that we had an anomaly that affected the CBOE's dividend futures contracts for 2017-Q1 (DVMR) and 2017-Q2 (DVJN). On Monday, 27 February 2017, the value of 2017-Q1's dividends per share increased by 3.5 cents to $12.70 per share, while the value for 2017-Q2's dividends per share decreased by the same amount to $11.83 per share. The anomaly then lasted until Thursday, 2 March 2017, when the gap between the forecasts for both quarters closed up once again. You can see the effect that these seemingly small changes had upon our dividend futures-based forecasting model for the trajectories shown for both 2017-Q1 and 2017-Q2 during the past week in the chart above.

We suppose that should teach us to not complain about the two values having otherwise been nearly identical since mid-January!

We recognize this change as an anomaly, first because it soon evaporated within a matter of days, and second, because stock prices didn't respond as our theory would suggest for investors suddenly shifting their forward looking focus from 2017-Q2 to 2017-Q1 during the last week. Had it been the real result of a significant change in expectations for the future for dividends, the upward movement for 2017-Q1 coupled with the sudden investor focus on the quarter would have sparked a much larger rally in stock prices.

But because it was just an anomaly, the shift in focus from 2017-Q2 to the nearer term future of 2017-Q1 didn't cause stock prices to move very much on the week, where the jump upward that occurred on Wednesday, 1 March 2017 wasn't repeated over the following days.

Still, that very limited Lévy flight did spark some useful observations into the nature of the mechanics that can drive such large, sudden upward movements in stock prices [hint: it wasn't the reaction to President Trump's speech on Tuesday evening]. If you haven't seen Nigam Arora's description of what really drove the market's action on 1 March 2017 yet, put it on your reading list - it's absolutely essential reading!

As for what else happened during Week 1 of March 2017, here is our list of the more significant headlines that influenced investors during the week that was.

Monday, 27 February 2017
Tuesday, 28 February 2017
Wednesday, 1 March 2017
Thursday, 2 March 2017
Friday, 3 March 2017

Elsewhere, Barry Ritholtz recalls the positives and negatives for the U.S. economy during the trading week that ran through Friday, 3 March 2017.

That's enough for this week. In next week's edition, we'll explain what that red box sketched on the chart above signifies.

Labels: ,

March 3, 2017

Just in time for St. Patrick's Day, a brand new invention promises to completely change the experience of how we, as Americans, enjoy our Shamrock Shakes at McDonalds: the Suction Tube for Reverse Axial Withdrawal!

Mere words cannot do justice to this kind of innovation, called simply "the STRAW", so let's turn to the promotional YouTube video that now accompanies all great technological achievements instead of just mere press releases (HT: Core77)....

Yes, it's real, and it's spectacular! Unfortunately, we couldn't find any U.S. patents assigned to McDonalds for its latest invention, and in fact, the closest we could find was U.S. Patent Number 9,627.095, which was issued to Solomon Kim on 27 December 2016, which describes how a long, flexible suction tube with additional holes at strategic locations along its length, might be deployed for the purpose of more completely extracting a fluid from the bottom of a pump-top container when the fluid level is very low.

But the patent for "the STRAW" would be a patent that we'd like to see!

Other Stuff We Can't Believe Really Exists


About Political Calculations

Welcome to the blogosphere's toolchest! Here, unlike other blogs dedicated to analyzing current events, we create easy-to-use, simple tools to do the math related to them so you can get in on the action too! If you would like to learn more about these tools, or if you would like to contribute ideas to develop for this blog, please e-mail us at:

ironman at politicalcalculations.com

Thanks in advance!

Recent Posts


This year, we'll be experimenting with a number of apps to bring more of a current events focus to Political Calculations - we're test driving the app(s) below!

Most Popular Posts
Quick Index

Site Data

This site is primarily powered by:

This page is powered by Blogger. Isn't yours?

Visitors since December 6, 2004:

CSS Validation

Valid CSS!

RSS Site Feed

AddThis Feed Button


The tools on this site are built using JavaScript. If you would like to learn more, one of the best free resources on the web is available at W3Schools.com.

Other Cool Resources

Blog Roll

Market Links
Charities We Support
Recommended Reading
Recommended Viewing
Recently Shopped

Seeking Alpha Certified

Legal Disclaimer

Materials on this website are published by Political Calculations to provide visitors with free information and insights regarding the incentives created by the laws and policies described. However, this website is not designed for the purpose of providing legal, medical or financial advice to individuals. Visitors should not rely upon information on this website as a substitute for personal legal, medical or financial advice. While we make every effort to provide accurate website information, laws can change and inaccuracies happen despite our best efforts. If you have an individual problem, you should seek advice from a licensed professional in your state, i.e., by a competent authority with specialized knowledge who can apply it to the particular circumstances of your case.