The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
No, Black Business Ownership Didn't Increase 400% in One Year
PolitiFact has the details -- but in any event, the purported statistic was suspect on its face.
Last week, I saw a report that seemed enheartening but left me suspicious -- to quote one account, "According to the Minority 2018 Small Business Trends survey, the number of black-owned small businesses in the U.S. increased by a staggering 400% in a year-over-year time period from 2017 to 2018." This statistic was picked up by a variety of sites, including conservative/libertarian sites, publications aimed at black readers, and general business sites. Last Monday, I e-mailed the publisher (Guidant Financial) to ask for details, but didn't hear back, and I was traveling with family so I put off further inquiries until today.
But PolitiFact (Kyra Haas) beat me to it; here's a quick excerpt:
In November 2017, small business financing company Guidant Financial and online credit marketplace LendingClub Corporation sent out an email survey to their clientele nationwide; they received responses from about 2,600 "current and aspiring entrepreneurs." The companies both posted general articles about the results of that survey on their websites in January without making any distinctions about race or ethnicity except that 47 percent of "aspiring entrepreneurs" surveyed were minorities.
The 400 percent increase in black-owned businesses claim, however, was not made until a subsequent graphic was published on Guidant's website on Aug. 8.
It was not clear from the released results how the companies came up with the 400 percent jump. The percentage of African-American respondents who owned businesses was not published, nor was that figure from last year.
The survey sample was not random, the response was voluntary, and the margin of error was not made public….
PolitiFact contacted Guidant about the claim in the graphic on Aug. 21. Two days later, a new version of the graphic was posted, this time without any mention of a 400 percent increase.
"We have found that the 400% statistic is being misrepresented in its context and therefore we have decided to pull that statistic from our info graph," public relations consultant for Guidant Stacia Kirby said in an email….
There's thus no real reason to think that the 400% increase claim is correct -- and ample reason, regardless of your politics, to doubt it, simply because 5-fold increases in this kind of number in one year are just so wildly improbable. Even if you think President Trump's economic plan is stupendous, or, if you prefer, President Obama's economic plan is stupendous and is just bearing fruit, a change like this doesn't happen in one year.
Recall that the claim wasn't even about a 5-fold increase in new business formation, or, if you prefer, a 4-fold increase in new business formation misleadingly reported as a 400% increase -- the claim was about such an increase in the total number of black-owned businesses. Even rates of change don't generally increase so sharply; but the total amount of pretty much anything almost never increases that way in a large economy (setting obvious exceptions such as the amount of newly developed technological goods, or the number of people in some subcategory of a group that itself sharply increased as a result of a sudden burst of immigration).
More specifically, the Census Bureau reported that "The number of black or African American-owned firms grew 34.5 percent between 2007 and 2012 — from 1.9 million to 2.6 million in 2012." That's a healthy growth, a 6% yearly increase. Is it really likely that the number grew 400% instead in one year? Or assuming that the number held more or less constant from 2012 to 2016, that it grew by 10 million in one year after that? (The total black population in the U.S. is about 43 million, and only a small percentage of any ethnic group owns businesses.)
So whenever you hear a claim such as that one, your Spidey sense should start tingling. It's not just that it's good to be true, it's too big to be true.
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
This misuse of percentages has always annoyed me, and I have seen enough misuse that I no longer believe them. If something went from 100 widgets to 110, go ahead and report a 10% increase. But if it increased from 100 widgets to 210, don't report a 110% increase, because I won't believe you; there have been far too many reports of 210% increase for me to take you literally. Say "more than doubled" and give the actual numbers to hang my hat on.
Seems like any proper style guide would forbid percentages of 100% or more.
I also resent the continual mixing of numbers and percentages. It does me no good to report "exports increased by $100M" because I don't know if that is 1% or 100%. Worst is reporting percentages for one aspect and amounts for another: "California increased olive oil production by 23% while Italy's dropped by 110 tons". I get that you have a thesaurus because you get tired of reporting that some murder victim is "attractive". But I don't see you switching between feet, yards, meters, and miles just to get some diversity, and those are all easily convertible. Amounts and percentages are not convertible.
Both raw figures and %ages serve their (mis)-purposes.
For example:
Under Obama* the SP500** went from about 805 to 1065, an increase of "only" 260 points but a percentage increase of about 32.25%***. For the same period under Trump the index rose from 2271 to 2896, a "whopping" 625 points but a percentage increase of "only" about 27.5%***. Which statistic more properly demonstrates market performance? Kinda depends on who one wants to lionize.
*The time frame used was inauguration date to 8/27 of the following year. A better metric might be from beginning of their first fiscal year to 8/27 of the following year but I'll leave that as an exercise for the reader.
**Data retrieved from Yahoo Finance.
***All math errors are the fault of the Deep State, the alt-Right or the Illuminati.
As the old adage states: There are lies, damned lies, and statistics!
It is unfortunate that journalists are generally innumerate.
If they were numerate, they would have been able to find real jobs. 🙂
It is unfortunate that their reader are so often innumerate.
>ahem< ...and so seldom read and edit what they have written, >blush
Bogus statistics appear almost daily and get repeated endlessly by the American media. The problem will only get worse until the media resolves not to publicize any more purported statistics until they have first been vetted by disinterested statisticians. That, however, is extremely unlikely to happen because the media itself seems to have lost all interest in providing balanced news coverage and instead only interested in taking sides. And, beyond that, startling statistics continue to sell newspapers and increase viewership, even when they are obviously bogus.
"There's thus no real reason to think that the 400% increase claim is correct -- and ample reason, regardless of your politics, to doubt it,..."
Whether the claim was a deliberate misrepresentation of the data, too subject to misinterpretation or mere innumeracy matters (in effect) little; the meme is out there and has been and will continue to be embraced in the homotextual world of the alt-Right. "The moving finger writes..." and all that.
Haven't seen (or specifically looked for) any announcement by Trump, et al, touting this claim but back in late 2017 Trump falsely claimed that black home ownership hit a historic high so I wouldn't be too surprised to see him also taking credit for this "historic rise."
An oldie but a goodie. One of the best purchases this math-challenged person has ever made.
http://a.co/d/c3lNslk
Bogus statistics appear almost daily and get repeated endlessly by the American media.
Another example: No, one-third of African Americans don't support Trump. Not even close.
From the pre-election polling and strong current social stigma against supporting Trump in any way (even a grudging "he's not doing too terrible" is enough to get you attacked in some quarters), I don't trust any opinion poll outside of the ballot box.
Uh, the pre-election polling was accurate. It showed Trump trailing Hillary by several percentage points... as did the ballot box.
But wasn't terribly accurate about where they'd get the votes, IIRC.
More specifically, the Census Bureau reported that "The number of black or African American-owned firms grew 34.5 percent between 2007 and 2012 ? from 1.9 million to 2.6 million in 2012."
That's a 40% increase. It sounds like 400% was just a typo.
You're conflating statistics from different sources (Census Bureau vs Guidant) and for different time periods (2007-2012 vs 2017-2018). The chart is gone from the website and no explanation as to what it purported to show seems to be forthcoming.
Exactly -- a 6% increase per year is plausible; a 35% increase in one year would be remarkable and a bit suspicious; a 400% increase in one year (which is what Guidant was said to have found) is implausible.
As I frequently say, torture statistics enough and they confess to anything. (Enhanced Interrogation and Statistical Methodology Correlation Studies in Criminology, Cook 2004)*
*sarc