Category Archives: junk science

Are Turkeys Calling a Market Top?

It's Thanksgiving Week 2017, and here at Political Calculations, that means that we'll be devoting the whole week to exploring the centerpiece of this uniquely American holiday in keeping with our annual traditions.

But that doesn't mean that we won't be discussing things like the stock market either - we'll just be loading it up with a healthy serving of Thanksgiving turkey.

Let's get started by doing just that, where we'll update the chart showing our favorite spurious correlation of all time - the apparent relationship that exists between the average live weight of U.S. farm-raised turkeys and world stock prices. The following chart updates that relationship through this point of time in 2017!

Spurious Correlation: Average Live Weight of U.S. Farm-Raised Turkeys and Annual Average of MSCI World Stock Market Index, 1970 - 2017(YTD)

If you're the type of person who believes that they can divine the future from any synchronous patterns you identify on charts showing apparently strongly correlated data like this - and the correlation here is indeed strong with an R² of 0.9717 - you should be very worried about the potential for a stock market crash in the near term, seeing as the chart shows that every time that the MSCI world stock index has risen higher than the proportionately scaled average live weight of U.S. farm-raised turkeys, a major correction hasn't been far behind.

We'll also tell you that even after detrending the data to account for the rising linear trends for both data series, the correlation doesn't disappear as you might suspect would happen for a fully spurious correlation. Instead, the R² drops to 0.5315, which might be considered to be a moderately-strong correlation.

And yet, we're happy to confirm that the apparent correlation is genuinely spurious. It's total garbage - the growth of the average weight of turkeys raised on farms in the U.S. is not, in any way that we can identify, connected to the growth of global stock prices.

If you want proof of how worthless this apparent relationship is, just consider that we first featured the spurious correlation between the average live weight of U.S. farm-raised turkeys and global stock prices back in 2014 - and as yet, the major sustained correction in stock prices that would seem to be imminent from this apparent relationship has not occurred.

So when you see charts showing these kinds of seemingly-correlated relationships, take them with a strong grain of salt! You should, at the very least, be able to identify some connection that logically links the two data series being compared. Without such a connection, you're likely just looking at something that, while it might be fun to consider, probably doesn't have much bearing upon or connection to the real world.

Speaking of which, since we've opened the door, if you're reading this article on a site that republishes our RSS news feed that also allows comments, plese share your links to examples of fun-but-false correlations. At the very least, the exercise might help you avoid the social disaster minefield that you would find yourself in if you're foolish enough to start talking up politics at this year's Thanksgiving feast by arming you with better and more interesting discussion topics.

Update: If you like puzzles and would like to take on an extra challenge over the holiday, you might consider looking into our detrending observation, where the trends that need to be subtracted from both series to properly detrend them are perhaps not linear ones!

Little Black Book of Junk Science

Little Black Book of Junk Science Cover -

The American Council on Science and Health has published a guide to many of the various kinds of junk science that often makes the news: The Little Black Book of Junk Science!

The book was written by the ACSH's Dr. Alex Berezow, who contributed two of the examples that we covered as part of our Examples of Junk Science series that primarily focused on pseudoscience in finance, economics and social sciences.

The Little Black Book of Junk Science however focuses more on where junk science is to be found in the fields of nutrition, biology, medicine and chemistry, which are the fields where the ACSH's staff has expertise. Here's a quick sampling what we thought were some of the book's more fun-related content:

Vaginal steaming

Endorsed by actress Gwyneth Paltrow, vaginal steaming claims if a woman sits over steaming water made with certain herbs, it will balance her hormones and help her uterus. A review declared it "sorcery for your vagina."


The reason your toilet doesn't work is because our government passed a law restricting flushes to 1.6 gallons each. Junk science claims that this is necessary to conserve water, even though water is not a scarce resource in most of the United States. It's also recycled efficiently in our sewage systems.

Genetic ancestry

Genetic ancestry tests use a small dash of science and a heaping scoop of speculation. Though your DNA contains information about your biogeographic ancestry, some commercial genetic ancestry tests may be little more than horoscopes. If you want to predict whether you will like cilantro or not, they are fine.


Earthing is the belief that running around barefoot somehow connects you to Earth's energy, which will improve your health. The exact opposite is true. We invented shoes because it protects our feet from injury and infection. Tell the neighborhood hippies to put their sandals back on.

Colon cleansing (Colon hydrotherapy)

Liquids go in your mouth, not in your butt. Unless you're constipated, there is no reason to give yourself an enema. Your body naturally detoxifies itself.

You can get the book in one of two ways. You can download a PDF version of the book for free or you can buy a physical copy from Amazon. Of the two, we'd recommend the physical copy, just so you can leave it sitting out where that special someone you know who can use this kind of information will see it!

The Most Surprising Finding of the Seattle Minimum Wage Study

Seattle Mayor Ed Murry Signs Seattle Minimum Wage Ordinance, 3 June 2014 - Source:

On April Fool's day in 2015, the minimum wage in the City of Seattle was increased by city ordinance from $9.47 per hour (130% of the federal minimum wage of $7.25 per hour) to $11.00 per hour (152% of the federal minimum wage), which would be followed by two additional minimum wage hikes within the next two years.

The second minimum wage hike mandated by Seattle's 2014 city ordinance took place on New Year's Day in 2016, when the same city ordinance mandated that the minimum wage at businesses with more than 500 employees in the city rise to $13.00 per hour (179% of the federal minimum wage), while small employers were required to increase the minimum wages that they pay to $12.00 per hour (166% of the federal minimum wage). On New Year's Day 2017, Seattle's minimum wage was hiked once more for the city's largest employers to $15 per hour (207% of the federal minimum wage), while small businesses were required to pay at least $13 per hour (159% of federal minimum wage).

As anybody with common sense might reasonably predict, mandating such a series of increases in the cost of labor for a business in such a short period of time without also mandating increased revenues to support it would likely lead to reductions in the amount of labor consumed by the businesses affected by the ordinance. And in fact, a new NBER paper authored by a team of University of Washington economists who had unique access to the payroll data of affected business found exactly that.

But that's the expected result from the analysis. What was surprising was that the authors used the highly comprehensive data set to which they had access in a successful attempt to replicate the results of one of the most controversial minimum wage studies on record: the 1994 Card-Krueger case study of the relative effect of a minimum wage increase upon employment in the fast food industry in adjacent communities in New Jersey and Pennsylvania.

This paper examines the impact of a minimum wage increase for employment across all categories of low-wage employees, spanning all industries and worker demographics. We do so by utilizing data collected for purposes of administering unemployment insurance by Washington’s Employment Security Department (ESD). Washington is one of four states that collect quarterly hours data in addition to earnings, enabling the computation of realized hourly wages for the entire workforce. As we have the capacity to replicate earlier studies’ focus on the restaurant industry, we can examine the extent to which use of a proxy variable for low-wage status, rather than actual low-wage jobs, biases effect estimates.

We further examine the impact of other methodological choices on our estimates. Prior studies have typically drawn “control” cases from geographic regions immediately adjoining the “treatment” region. This could yield biased effect estimates to the extent that control regions alter wages in response to the policy change in the treatment region. Indeed, in our analysis simple geographic difference-in-differences estimators fail a simple falsification test. We report results from synthetic control and interactive fixed effects methods that fare better on this test. We can also compare estimated employment effects to estimated wage effects, more accurately pinpointing the elasticity of employment with regard to wage increases occasioned by a rising price floor.

Our analysis focusing on restaurant employment at all wage levels, analogous to many prior studies, yields minimum wage employment impact estimates near zero. Estimated employment effects are higher when examining only low-wage jobs in the restaurant industry, and when examining total hours worked rather than employee headcount.

What makes their success in replicating the results of the Card-Krueger study by filtering the Seattle data to reproduce its limitations is significant in that it effectively invalidates Card and Krueger's 1994 finding that minimum wage increases have no effect upon employment. Simply put, the limited nature of the data that Card and Krueger used to support their earlier study of the effect of New Jersey's 1992 minimum wage hike almost certainly led them to miss its true effect on employment after it went into effect.

Source: /

This same issue of data detail has come up before with economists who rely upon income tax data to measure income inequality, which similarly fails to capture the true nature of the distribution of income by not providing the additional individual-level detail that other data sets provide. We've described the knowing use of such limited data without acknowledging its limitations as "analytical malpractice", which in the worst cases, crosses the ethical line into outright pseudoscience.

To be fair, we believe that the Card and Krueger's case study was a good faith effort that applied a novel approach to attempt to measure the impact of a minimum wage hike on employment. The limitations of the data they had available however meant they were weren't capable of detecting the reduction in labor hours that occurred across all employees in the industry, which the Seattle minimum wage study indicates would have negatively affected many whose wages are above the levels that would be directly impacted by minimum wage increases.

Since we've touched on the topic of pseudoscience in this post, particularly where the limitations of data are concerned, we should note that there's more going on with respect to the analysis of the impact of Seattle's minimum wage hikes that more strongly fits into that category. Specifically, as Jonathan Meer has observed:

This paper not only makes numerous valuable contributions to the economics literature, but should give serious pause to minimum wage advocates. Of course, that’s not what’s happening, to the extent that the mayor of Seattle commissioned *another* study, by an advocacy group at Berkeley whose previous work on the minimum wage is so consistently one-sided that you can set your watch by it, that unsurprisingly finds no effect. They deliberately timed its release for several days before this paper came out, and I find that whole affair abhorrent. Seattle politicians are so unwilling to accept reality that they’ll undermine their own researchers and waste taxpayer dollars on what is barely a cut above propaganda.

That sounds startingly similar to the "battle of the experts" dynamic described by former antitrust litigator David Gelfand in our Examples of Junk Science series, which we should note also fails the Goals, Progress, Challenges, Inconsistencies, Models and Falsifiability categories in our checklist for detecting junk science.

Does Dividend Acceleration Predict Real GDP?

We've known since 2008 that the acceleration of dividends can be used to better explain how stock prices behave, but we've never really explored whether they can also tell us about the relative health of the U.S. economy.

Until now! Hypothetically, we should see some sort of correlation between changes in the rate of dividend growth for large, publicly-traded firms and the growth rate of real GDP in the U.S., where in times of decelerating dividend growth (or negative acceleration), we should see low GDP growth rates, and in times of positively accelerating dividend growth, we should see higher levels of real GDP.

Alternatively, the null hypothesis would be that there's not a significant or strong enough correlation between the two for that kind of exercise to be worthwhile.

Today, we're going to do a quick, first pass analysis using a limited set of data to see how worthwhile it might be to explore the hypothesis further. We'll use our real GDP temperature chart for data going back to the first quarter of 2000 to divide the annualized one-quarter inflation-adjusted GDP growth rates that the BEA reports each quarter into "cold", "cool", "moderate", "warm" and "hot" levels of real GDP growth.

One Quarter Growth Rate for U.S. Real GDP, 2000-Q1 through 2017-Q1 First Estimate

In our next chart showing the change in the year over year growth rate of the S&P 500's trailing year dividends per share, we've indicated the periods where real GDP growth dropped into the "cold" range, where real economic growth in the U.S. was less than or equal to 1.5%. This visual presentation should quickly tell us if there's any kind of basic correlation.

Change in Year Over Year Growth Rate for S&P 500 Trailing Year Dividends per Share, January 2000 through April 2017, with Dividend Futures through December 2017

Looking over this second chart, we have 14 periods of one longer or more where the real GDP growth rate dropped into the cold zone. We also see 14 periods where the change in the rate of growth of the S&P 500's trailing year dividends per share dropped into negative territory.

However, we see that the two sets of negative occurrences don't align particularly well, where we have some periods with negative dividend growth acceleration without corresponding cold GDP levels, and vice versa. This visual correlation check tells us that there's not necessarily a particularly strong correlation between the two sets of data that might be worth evaluating further, where the next step is to note our results so that others can take them into account in shaping their own work if they're engaged in similar analysis.

Sure, we could follow the example of a shockingly high percentage of medical and psychological studies fail to be replicated because the authors might have p-hacked their way to statistically significant outcomes that they could publish. We however find it more useful to publish the path that led to the apparent dead end, because while not likely to ever be picked up by any journal, we can at least save others a lot of time by identifying what either doesn't work or what approaches don't appear offer much in the way of promise for either practical application or for future work.

That's the very real economic value of publishing results that fail to uphold the main hypothesis (or rather, that fail to reject the null hypothesis). It's nowhere near as glamorous as selectively publishing the kind of results that might get trumpeted throughout the media, but often, it's much more valuable.

Separating Science from Pseudoscience

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!