Political Calculations
Unexpectedly Intriguing!
August 26, 2016

Several weeks ago, Political Calculations launched a new series, "Examples of Junk Science", where we are specifically focusing on where flawed analysis and the misuse of data leads to the spread of false or misleading information in the fields of investing, finance and economics.

More than that, we're also seeking to present explanations for why the examples of flawed analysis and misuse of data that we present lead to results that are not valid, which we hope will be beneficial for analysts who seek to make positive contributions in these fields, where the ultimate goal is be to improve the general quality of analysis that is produced altogether. We hope to do that in part by highlighting real examples so you can more quickly recognize them as the pseudoscientific junk they are so you can directly challenge the people who propagate them.

Today's example of junk science is representative of what may quite possibly be the most common way that analysis is twisted to present a misleading picture of the economy. Coincidentally, it is also perhaps the simplest: stating changes in values as a percentage of a percentage.

This is a technique that can greatly exaggerate the significance of a claim, which often will fall apart when put into a more appropriate and relevant context. As such, this kind of data mishandling falls under the precision category of the checklist for how to detect junk science.

How to Distinguish "Good" Science from "Junk" or "Pseudo" Science
Aspect Science Pseudoscience Comments
Precision If numbers are presented in support of a scientific explanation, they must be stated with the precision and accuracy required by their level of significance as determined by known measurement error in the data from which are derived, neither more nor less. Pseudoscience practitioners will often present numbers with a level of precision and accuracy that exceeds that supported by the known accuracy of real world data in order to give the appearance of greater validity for their claims. A recent example of pseudoscientific deception by precision include certain economists suggesting that "a Keynesian multiplier of 1.57" specifically applies for government stimulus spending, when a wide range of studies suggest the actual multiplier may be "anywhere from 0 to 1.5" (note the difference in the number of decimal places and potential range of values!)

Today's example isn't so much about decimal places as it is representative of the kind of deception that results when precisely calculated values are placed into an inappropriate context. Let's go over the details....

A chart similar to the one below was included in a blog post under the heading “Bank C&I Loan Charge-Offs Soaring Again”. This chart caught my attention because it seems to indicate that bank C&I (Commercial and Industrial) loan charge-offs are happening at one of the fastest rates of the past 30 years — the sort of rate that would be consistent with the US economy being in recession.


The problem is that the above chart shows the percentage change of a percentage, which opens up the possibility that what is in reality a small increase is being made to look like a large increase. For example, an increase from 1% to 2% over the course of a year in the proportion of loans charged-off would be a 100% increase if expressed as a year-over-year percentage change in the percentage of charge-offs, whereas all you’ve actually got is a 1% increase in the total proportion of loans that have been charged-off.

The next chart is based on exactly the same data, but instead of displaying the year-over-year percent change in the percentage of C&I loans that have been charged off it simply displays the percentage of C&I loans that have been charged off. This is not just a more correct way of looking at the data, it is a way that has not given any false recession signals over the past 30 years.


The first chart’s message is: an economic recession is either in progress or imminent. The second chart’s message is: the US economy is not in recession and is presently not close to entering recession.

The same data, opposite messages.

No matter how you slice it, the presentation of a percentage change of a percentage is always highly misleading, even if mathematically correct. At the very least, it is more than one step removed from the actual numbers used to calculate the original percentage, which is the appropriate context in which the percentage can be understood.

Believe it or not, the use of percentage changes of percentages in place of the more accurate and direct presentation of percentage changes of numbers can have legal consequences, as Republican members of Arizona's state legislature discovered in 2008.

Judge orders rewrite of sales tax analysis

PHOENIX — A state judge ordered late Friday that the description of a proposed tax increase for transportation be rewritten to exclude a calculation of how much the levy will increase.

The description is to be put into pamphlets mailed to the home of every registered voter.

Maricopa County Superior Court Judge Edward Burke said it may be "mathematically correct" to say that Proposition 203, which would boost state sales taxes from 5.6 cents on every dollar spent to 6.6 cents, equals a 17.8 percent increase in what people will pay in sales taxes.

But the judge said such calculations "are likely to mislead many voters."

He accepted the arguments by Paul Eckstein, the attorney for tax backers, that the figure was deliberately inserted by Republican lawmakers to convince voters to reject the levy.

Mike Braun, who represents the Legislative Council, told Burke the law requires the panel to not only explain each ballot measure but also how approval would affect existing law.

He said it would be "helpful" to tell voters that whatever they are paying in state sales taxes would increase by 17.8 percent if Proposition 203 is adopted.

Eckstein, however, told the judge the accuracy of the calculation is beside the point.

"An analysis can be 100 percent accurate, and it can be unfair and misleading," he said.

Burke agreed.

For example, he said, raising the tax rate from one cent per dollar to two cents has an "absolute percentage increase" of 1 percentage point. But it has a "relative percentage increase" of 100 percent.

"Many voters are likely to confuse relative and absolute percent increases," the judge wrote.

As you can see, the components for the state sales tax example has a direct parallel component in the loan charge-offs example. The problem, as described by the judge in this case in non-mathematical terms, comes when the absolute percentages represented by the math above, the percentages of numbers, are confused with relative percentages, the percentage changes of percentages. The presentation of the relative percentages is misleading because their calculation is too many steps removed from the original base values from which the original percentages were calculated, which breaks the link to the specific context to which they are relevant.

In the case of today's example of junk science, it's the difference between suggesting that the risk of recession is imminent or of there being a comparatively low risk of recession in the near term. Given how different those outcomes are, expressing calculated values in the correct context is essential to avoiding the consequences that would come from taking the wrong actions in response to misleading information.


Saville, Steve. You can make statistics say whatever you want. The Speculative Investor. 19 July 2016. Republished with permission.

Political Calculations. How To Detect Junk Science. [Online Article]. 19 August 2009.


August 25, 2016

With so much destructive flooding in Louisiana during the past week, we couldn't help but think back to this past June, when the region around Houston, Texas experienced a tremendous amount of rainfall that also led to significant flooding and lots of property damage.

But one resident in the town of Rosharon, Texas beat the flood, using some remarkable technology developed in Louisiana for the purpose of preventing property damage. The following video tells the tale.

At $8,300, the cost of the residential AquaDam may seem high, but for those who live in flood prone areas, where flooding can cause property damage that can exceed several hundred thousand dollars, it may be a pretty reasonable investment. And the best part is that with the right combination of pumps, homeowners could actually use the same water that might otherwise destroy the value of their homes to protect them.

The question of whether its worth the cost however comes down to a combination of how much risk there is of property damaging flooding and the impact of how costly it could be. To estimate if it is worth the cost, as a general rule of thumb, you can multiply the cost of potential property damage by the probability that flooding will damage the property over an extended period of time. If that number is greater than or equal to the cost of using technology like the Aqua Dam to mitigate against the potential for damage, then it makes sense to invest in the technological solution.

It's not quite that cut and dried however, because there is a lot to be said for peace of mind if that number only falls short by a relatively small percentage, where choosing to invest in technology to overcome such a risk would come down to a personal choice.

The same principles apply for adaptation to climate change, where one could argue that this kind of technology could be more cost-effectively applied to provide additional protection to a levee, which if overtopped by flood waters, could easily result in millions of dollars of damage to an entire community. That may be especially true when the alternative course of action may involve an even more costly project to build up the levee.

Then again, the cost of the kind of technology to deal with such hazards may be so prohibitive that property owners may simply choose to buy extra insurance protection to pay for what it costs to clean up and recover after the damage happens, if and when it does.

And that's the rub. It all comes down to placing a bet on the statistical likelihood of something bad happening, which may or may not when it matters most to you. How much are you willing to wager to win on a choice like that? And how much are you willing to risk losing if you fall on that other side of the bet?

Update: Matt Kahn applies a little more complex math than what we described above to identify what kinds of people would choose to live in areas prone to flooding in the absence of government subsidized insurance protection (and by extension, since the math would be similar, technological protection).

Labels: ,

August 24, 2016

Every three months, just about at the time that the market's quarterly earnings season ends shortly after the midpoint of the quarter, we take a snapshot of the expectations for future earnings in the S&P 500.

Today, for the first time in the last seven quarters, the trailing twelve month average of the S&P 500's earnings per share has risen, which suggests that the earnings recession which took hold in the fourth quarter of 2015 may finally have bottomed.

Forecasts for S&P 500 Trailing Twelve Month Earnings per Share, 
2011-2017, Snapshot 2016-08-18

Going by S&P's current projections, the index' trailing year earnings per share may return to its pre-earnings recession levels in the first quarter of 2017.

The improvement in the S&P 500's earnings may be directly attributed to the rebound of crude oil prices since they bottomed in early February 2016, which have eliminated the downward pressure on earnings that have prevailed in the market's oil production sector since the end of June 2014.

Through Monday, 15 August 2016, the price of West Texas Intermediate crude oil at the Cushing, Oklahoma terminal stood at $45.72 per barrel, a nearly 75% improvement over the $26.19 per barrel it bottomed at back on 11 February 2016. Oil prices would have to rise by another 136% from that $45.72 per barrel figure to return to the $107.95 per barrel that was recorded on 20 June 2014.

We think that for S&P's latest projection of future earnings to hold, oil prices would have to continue their overall rise. However, since the current quarter of 2016-Q3 began, those prices have been somewhat choppy, so we think it is more likely that the recovery of the S&P 500's earnings per share to levels back above their pre-earnings recession levels will take longer than S&P projects.

This increase in the S&P 500's trailing year earnings per share is the first real positive news that the market has seen for its earnings prospects since mid-2014, and particularly the oil producing sector of the market. After having fallen for six consecutive quarters, it's about time that industry had some good news show up on its bottom line.

Data Source

Silverblatt, Howard. S&P Indices Market Attribute Series. S&P 500 Monthly Performance Data. S&P 500 Earnings and Estimate Report. [Excel Spreadsheet]. Updated 12 May 2016. Accessed 19 August 2016.

Labels: ,

August 23, 2016
Source: U.S. Department of Defense - http://www.defense.gov/Media/Essay-View/CollectionId/15653

InsureBlog's Hank Stern reached out to us last week in connection with an effort to link blogs and other social media outlets with support for much needed disaster relief in the flooding-struck Gulf Coast region of the U.S. We're going to keep the ball rolling by pointing to a number of disaster relief resources that should draw your attention no matter where in the world you are.

With much of the worst flooding having taking place in Louisiana's Acadiana region, a great place to start identifying ways to make the biggest difference with your donation or your volunteer efforts can be found through Baton Rouge's CBS affiliate WAFB's site, which has been regularly updating a list of needs and volunteer opportunities in the region.

We've also selected a handful of charitable organizations that are focused on providing disaster relief across Acadiana, which are tied to national organizations that have been ranked highly by Consumer Reports and the Christian Science Monitor for their effectiveness in putting donated funds into real relief efforts.

American Red Cross

The American Red Cross has channels to either support taking your donations to support their efforts or to direct you to where hands-on volunteers are needed, including for blood donations.

United Blood Services

One of the things we learned during the aftermath of Hurricane Katrina is how rapidly available blood supplies across the nation can dwindle because of the disruption of blood donations in the disaster-affected area of Louisiana and the Gulf Coast. Going to your local Red Cross or UBS blood donation center or participating in a blood drive can do a lot to alleviate a potential shortage that would now in the making.

Catholic Charities Archdiocese of New Orleans

Much of the flooded area falls within the parishes served by the archdiocese of New Orleans, which has a well established distribution network to provide disaster relief services in this area.

Salvation Army

The Baton Rouge, Louisiana Corps of the Salvation Army is actively engaged in providing assistance to families displaced by flooding and is also very much in need of volunteers to support its efforts.

The United Way

There are two branches of United Way that are well positioned to provide disaster relief: United Way of Acadiana and United Way of Southeast Louisiana, both are accepting online donations.

We would be remiss if we didn't recognize how social media is making disaster relief efforts more effective. To that end, we're really impressed by an organization that didn't exist during the days of Hurricane Katrina, but which has already made a big difference in its short history: Crisis Cleanup. Here's a good description of what they're about:

How do you direct 30,000 volunteers from 100 organizations to 5,000 locations across a 500-mile arc in 8 weeks?

You don't. You let them organize themselves.

Crisis Cleanup empowers relief organizations to instantly coordinate response efforts in real time, redirecting thousands of volunteers from to waiting lines to survivors' basements.

Crisis Cleanup was developed by and for field volunteers, team leaders, canvassers, and the people who work one-on-one with survivors whose homes have been affected by flood, tornadoes, earthquakes, wind, fire, or other disaster. Crisis Cleanup can respond to a new disaster the same day, permitting relief organizations to instantly coordinate efforts.

Crisis Cleanup implements a "Craigslist" philosophy to recovery efforts– organizations that are aware of work orders enter them into the system, and organizations with capacity to help can claim and perform the work. The system is not public, but it is open and transparent among participating organizations. No centralized organization is "in charge." This non-threatening approach minimizes duplication and maximizes communication, coordination, and efficiency.

While entering a client into Crisis Cleanup does not guarantee that he or she will be served, it guarantees visibility and maximizes the chances for assistance, while helping relief organizations prioritize their limited resources.

It's been over 10 years since the flooding from Hurricane Katrina demonstrated the need for this kind of decentralized coordination tool for making relief efforts more effective. We're excited to see this particular evolution in how disaster assistance is directed.


August 22, 2016

August 2016 is going to be a long month. We're three weeks into it with two more to go and already, with no sign as yet the summer doldrums in the S&P 500 are anywhere near ready to give way to volatility.

Not that there's anything wrong with that - as an investor, a not-so-volatile, yet almost painfully slow general rise in stock prices is a good thing. Especially as compared to the fireworks that created the echo effect in our standard model's projections of the S&P 500's stock prices through the rest of the quarter.

Alternative Futures - S&P 500 - 2016Q3 - Standard Model - Snapshot on 2016-08-19

The echo effect is the result of our futures-based model's use of historic stock prices as the base reference points from which we project the alternative trajectories that the S&P 500 is likely to follow into the future based upon how far forward in time investors are looking in setting their expectations when making their current day investment decisions. In this case, the echo effect began wreaking havoc with our standard model's projections on Thursday, 18 August 2016, on the anniversary of the beginning of China's August 2015 stock market meltdown.

To try to improve the accuracy of our model during the periods where we know in advance that its projections will be less accurate as a result of the echo effect, we've substituted the average daily trajectory that the S&P 500 has taken through this period of the year, based on historic stock price data going back to 1950, which smoothes out the echo of last year's volatility. The following chart shows how well that worked during the first two days in which we have to compensate for the echo effect.

Alternative Futures - S&P 500 - 2016Q3 - Modified Model 01 - Snapshot on 2016-08-19

Overall, so far, so good, but it's way too early to make any sort of judgment call on how well this modified method will work out to be. At this point though, it is pretty safe to say that Week 3 of August 2016 looks like it ended as it began, not to mention how it behaved in between, with investors appearing to be fairly closely focused on the distant future quarter of 2017-Q2 and the expectations associated with it in setting their investment decisions during the past week.

Compared to last year, that's good news. We'll have to wait to see if the next week will bring any excitement. Until then, here is a selection of market-related headlines that caught our attention during the past week.

Monday, 15 August 2016
Tuesday, 16 August 2016
Wednesday, 17 August 2016
Thursday, 18 August 2016
Friday, 19 August 2016

For a listing of the week's positives and negatives as it relates to the markets and the economy, be sure to check out Barry Ritholtz' succinct summation of the week's events.

Labels: ,

About Political Calculations

blog advertising
is good for you

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.