Troll Poll

Via.

The Hill reports the Gallup Poll (and the Weekly Standard gets very excited):

Economic confidence has been holding at a seven-month low, according to a survey released by Gallup on Tuesday.

The Economic Confidence Index averaged -8 in June, which Gallup said is statistically tied with May’s -7 reading.

Gallup’s latest figures are keeping the index at the lowest monthly reading since November....

Gallup's Economic Confidence Index is the average of two things: the public’s rating of current economic conditions and whether people feel the economy is improving or getting worse.

Then again, according to the Conference Board (via Deseret News):

Consumer confidence showed a solid gain in June following a modest increase in May, supporting the view that strong job gains are giving a boost to the overall economy.

The Conference Board said Tuesday that its consumer confidence index rose to 101.4 in June, up from a May reading of 94.6. The June level matches the level in March. The index took a tumble in April.

The index is now 17.4 percent higher than it was a year ago, evidence that the economy is poised to enjoy stronger growth in coming months.

The June gain, which was stronger than economists had expected, was supported by increases in consumer views about current conditions and the labor market. Those expressing the view that jobs were plentiful rose to 21.4 percent, up from 20.6 percent in the previous month.

What's up with that?

One thing to note is that the Gallup numbers take zero as the baseline while the Conference Board takes 100, so that the one looks like a pendulum between bad (negative) and good (positive), while the latter seems more like an oscillation from mediocre (two digits) to super (three digits), which captures the sentiment better, I think (in practice, their zero seems to correspond more to about 100 in the Michigan Consumer Sentiment Index, so that it's almost always below the line unless Bill Clinton is president wherever you draw it, whereas that point is up around 120 in the cheerier Conference Board picture).

Is it the reality that's relentlessly negative, or the picture?.

Another is that the Gallup index appears to have been published only since February 2008, when it made its debut at a scary -36, on its way down in that year's crash to -65 (the chart above is of monthly averages, so the numbers are less extreme), so it's never seen positive territory at all in the course of its public existence until last December, and reached its all-time high of 6 last Valentine's Day. When it reached -7 last November, that was a 17-month high, since June 2013, and only the second time in its history it had gotten so high. So when it fell down to -7 again at the beginning of July it could just possibly not be the end of the world yet, as you can see if you look at the curve for the three-day rolling averages since a year ago:

Gallup Economic Confidence for the last 12 months   (July 2014–July 2015).

A more important question, probably, would be why the Gallup and Conference Board numbers are marching in opposite directions. Is one calculation better than the other? It's certainly true that the Gallup is a simple mean off of two factors and the Conference Board a more complex basket of data, including things consumers know quite a lot about (their own current economic situation and prospects) alongside the things they know as little about as economists do, possibly in some cases even less (how is the economy and where is it headed overall).

Conference Board Consumer Confidence; in the upper right the chart for the last 12 months.

Also, the Michigan Index data agrees pretty emphatically. As their chief economist Richard Curtin reads it:

Consumers voiced in the first half of 2015 the largest and most sustained increase in economic optimism since 2004. Just as important, that same record was set by households in the top third of the income distribution as well as by the middle third and those in the bottom third of the income distribution. Moreover, the recent surveys recorded those same records when consumers were asked to evaluate prospects for the national economy, their personal finances, and buying conditions.

Leaving the only real question as why do The Hill and the Weekly Standard only look at the outlier? And I guess it answers itself, huh? It's the one that gives them the results they're looking for,