Why you can’t sum up Obama in 9 charts
This meme has been circulating around the internet “proving” how Obama is ruining everything. Trump posted a version on Twitter. The Washington Post kinda fact checked it. It arrived in my inbox with the challenge “If you don’t believe it, just Google.” So I did.
And that took me right to the source.
In brief, there are many reasons why these graphs are either misleading or just plain wrong — even beyond the way the charts selectively crop off positive data.
Below is a non-comprehensive list of those issues.
I encourage anyone to look at the graphs themselves (links included!), compare them to other indicators, look at them over different time scales, and generally familiarize yourselves with the facts.
And for bonus points, keep this page open in another tab for deeper context.
This graph becomes more interesting when you overlay the percentage of the US population that has Bachelor or Higher Education Degrees:
which seems to indicate total student loans may be increasing partly because there are more students getting degrees.
However, it is no secret tuition is also increasing, but the reasons for this can’t be summed up in a chart.
(Side note: It appears the shared charts arbitrarily labeled the axis and filled in the time before 2006 (from when the FRED has stats) with zeros making that spike look HUGE! You could label this same chart as “Misleading Statistics Shared on the Internet” and fill everything in with zero before sometime in 1980, and boy will there be a spike!)
First, here is the full graph:
You’ll immediately notice, the internet meme charts fail to include the recent drop.
More interesting, however, is a chart of “change from previous year.” This gives a better look at how each year’s spending compared to the previous:
You clearly see that after the last recession (shaded areas on the charts), although the total amount paid out may have increased, the rate of increase each year has been sharply decreasing. This means the SNAP payouts have been steadily turning around, until total payouts began decreasing as seen in the above graph.
Another way of interpreting this data is by looking at the Participation Rates:
Participation Rate basically says, “Of all people who are eligible to receive food stamps, what percentage are actually using the program.”
It is conceivable that if there is an increase in eligible people using the program (possibly due to better outreach and advertising) — even if there is an overall decrease in eligible people — there may be an increase in money spent.
Another fun one. Has debt (as a percent of GDP as the graph shows) increased? Yes. When have those increases occurred?
This graph of annual change:
clearly shows that since 2009 (when Obama took office, and when the recession ended (grey shading)) the rate of debt increase has sharply decreased.
Overlay a similar view (total debt change per year) with the deficit (difference between what the gov takes in and spends):
and you see that since 2010, the budget has been getting closer to balanced (something we haven’t seen since Clinton’s administration) while the rate of debt increase is slowing.
(Side note: If we want to try to assign all the cause / effect of debt on the presidents (ignoring everything else happening in the world), lets look at the change in debt between when a president took and left office.)
The graph indicates Monetary Base which is not money printed, I think. It may have something to do with it — I really don’t know. I think it’s a measurement of all currency. Both in circulation, and held by banks and the reserve. It is complicated. But, rather than spend the time figuring out what that actually is (or if it means anything) FRED gives us two other sets of data, “Value of Bills Printed,” and “Money in Circulation”:
(Note: if someone can explain Monetary Base, how having more money in the banks (not in circulation?) affects the economy, or other ways this is relevant, please do so in the comments.)
Health Insurance Costs:
I can’t find the source for this graph (since it is not FRED), but here is some health related info from FRED:
The green line appears to be the data used for the graph… but feel free to find something more relevant.
The green line (personal expenditures) looks similar to the chart in the meme. However, the meme chart shows only a small period in time: roughly ’09 to ’13 which, sure, if you zoom the graph to that tiny period, things look really bad… but that totally ignores any sense of historical context.
Another detailed report by US News and World Report gives lots of health indicators, showing how complex this all is.
(Side note: And to get a the direct question of whether premiums have risen disproportionately under Obama…)
Labor Force Participation:
It is important to note that Labor Force Participation (LFP) is the number of people who are either A) Employed, or B) Not employed but are looking for a job. What does a decrease in percentage of the population not looking for work mean?
Partly, a pocket of ageing people (baby boomers) leaving the workforce and entering retirement. Did they give up looking for work and retire early due to the recession? Possibly some did.
But rather than quibble that point, lets look at actual employment rates (green), and unemployment rates (red) over time (the blue line is a squished version of the LFP graph seen in the meme):
Suddenly things looks a little different as we see unemployment rate has decreased steadily, while employment rate has increased steadily.
Workers Share of the Economy:
Here’s the full thing:
So this has been decreasing steadily since 2001 (GW Bush takes office) and now it is slowly starting to increase.
Here is the same graph with Real Median Personal and Household Income:
(Side note: These charts include CEO compensation as “labor” — interesting to note the wage gap between workers and CEOs. Which means that during all this, any gains are disproportionately going to upper-level management, not workers… but I digress.)
Median Household Income:
You’ll note the meme graph is cut off in 2012 (when things start to increase again) and starts in 2000. Here’s the full thing:
And here it is with actual dollars as well (the first is 2014 inflation adjusted):
And with personal income:
But even more interesting is how things are changing over time:
which shows a clear recovery — till 2013, anyway.
(Note: there doesn’t appear to be data from 2015 or 2016)
Again, historical context helps:
The other line shows the number of privately owned housing units that completed construction — which has been increasing steadily since 2011-ish.
So… we are back to 1995 levels. And when did the decline begin? (Anser: 2004.)
Who was president then? OK, I’m not trying to cast blame here, just pointing out that there are lots of forces at play that aren’t directly attributable to Obama.
Moving on, here is the home ownership rate with the Mortgage Delinquency Rate:
which shows a lot of bad stuff happening starting around 2006 / 2007.
In short, it is powerfully dismissive to try and sum up any administration with 9 charts. 16 helps get us closer — but it might be worth spending some time with this stuff.
When you look at what was happening before and after the recession, a different picture materializes than what 9 carefully cropped graphs would appear to indicate.
If nothing else, this exercise provides a good example of how proper scale of a chart can be used to clarify information — or obfuscate information.
Always go to the source.