How Truth is Represented in Economic Forecasting & Data Analyses | Blog#42

Jared Bernstein presents an aspect of data reporting in economic indicators that is not often discussed, but should be. I’ve blogged and re-blogged about this very issue in the past, and given recent polemics on the Sanders economic platform, this topic merits a re-visitation.

Bernstein interviewed Northwestern University economics professor Charles Manski for his piece in the Washington Post.  I can’t recommend enough that you click through to read the entire piece, which I cannot re-post here. The piece centers around how data are presented without any illustration of the uncertainty that should be built-in. That illustration, outside of the US, comes in the form of “error bands” around the data that are presented. Those bands provide the reader with a visual representation of what error might predictably look like. The example in Bernstein’s piece is from the Bank of England’s fan charts, but there are others.

What I would like to draw attention to the most, for the purposes of this piece, are these two quotes from Jared’s interview:

“A broad reason is that governments, firms and individuals use official statistics and forecasts when making numerous decisions. In the absence of error bands, they may misinterpret the information that the numbers provide. Naive users may incorrectly take central forecasts at face value. Sophisticated ones may be aware that the forecasts have errors but not have a sense of how large these errors may be.

As a result, the quality of decisions may suffer.”

and

“… And I have heard it from staff at the CBO when I have argued for expression of uncertainty when scoring legislation. I can’t prove that the fear is misplaced, but I worry that it ignores an opposing danger. That is, if government agencies continue to suppress uncertainty, they will lose credibility when people eventually wake up and recognize that official statistics and forecasts may have large errors.”

At a time when we have divided government and obstruction on the magnitude we’ve seen in the last six years, official data are all the more important in a polarized society. The degree to which we are polarized by far exceeds the political realm and reaches well-into the corporate mainstream media, with narratives that are intentionally recounted in less than complete ways, and data that are both incomplete and uncertain, being presented at face value and as fact.

I’ve written about the implications of the changing of the guard at the Congressional Budget office, whose responsibility it is to report on jobs numbers and unemployment, with the changing of the guard after Election 2014. These are the pieces I’ve written or reposted:

https://www.rimaregas.com/2014/10/fixing-the-downward-bias-in-the-unemployment-rate-jared-bernstein/

and

https://www.rimaregas.com/2015/08/jared-bernstein-on-our-economic-outlook-the-fed-unemployment-and-fedup-blog42/

and

https://www.rimaregas.com/2016/02/how-the-unemployed-are-counted-scoundrel-edition-msmbias-on-blog42/

To be fair to Kevin Drum, he has written an update to his opinion of Bernie Sanders’ economic prescription in, “On Second Thought, Maybe Bernie Sanders’ Growth Claims Aren’t As Crazy As I Thought.” That said, better presentation of data, with error bands as the standard, would force everyone, pundits included, to conform to a modicum of honesty.

There are still millions of people who never regained jobs that can be considered a viable replacement to those they lost during the mass-layoffs at the start of the Great Recession. There are millions of young Americans who have graduated each year since the start of this recession and have neither found the kinds of jobs their new degrees should lead to, or aren’t paid enough to cover basic needs, much less repay huge student loans. Black unemployment, the subject of the next installment in my precariat series, is still through the roof.

Wouldn’t it be fair if all of us could look at charts the government and the media put out and see for ourselves what the margins of error might look like? Wouldn’t it be nice, every time some government agency or Congress puts out a report, that the charts showed a band of optimism and pessimism around the data offered?

This is how it’s done in Europe for the sake of transparency. It’s not done in the U.S. because the truth might make us vote completely differently.

Take that to heart. Demand better. Expect better. You deserve better. You deserve to see the truth for yourselves and be trusted to evaluate and act accordingly.


Blog#42 is a reader-supported site and my full-time job. I cannot do more without your financial support. Please click here to help me take Blog#42 to the next level!


Recommended reading:

Everything Jared Bernstein ever posts on his blog.

Charles Manski, “Communicating Uncertainty in Official Economic Statistics: An Appraisal Fifty Years after Morgenstern.”

Tutorial: How to create error bands in Excel