A thought for Sunday: the most important issue in the 2016 election was . . .

 - by New Deal democrat

This is a post I've been meaning to write for several months. For a while after the election last year, there was a debate about whether the "economic anxiety" in the (white) working class was the most important factor vs. was it simply a matter of racism. The consensus has nearly settled on the narrative that racism was decisive, to the point where "economic anxiety" has become a taunt, and some who embrace identity politics actively disparage progressive economic issues.

I'm here to show you data that - in part - disputes that consensus. What was the most important issue in the 2016 presidential election?  The below data on that issue all comes from the Voter Study Group, from its survey published several months ago: "Insights from the 2016 Voter survey."  

In the below graphs, the potency of various issues are examined in terms of how well they lined up on a liberal/conservative or favorable/unfavorable axis, but for simplicity's sake it is pretty clear that they correlate with a vote for Clinton (left) or Trump (right).  The more vertical the line, the more decisive the factor, whereas a horizontal line means that the factor made essentially no difference in whether a vote was for one candidate or the other.  the 2016 results are in red, vs. the 2012 results in gray. What I've done is to delete the names of the nine factors they tested, so you won't be swayed by any pre-existing opinion you might have had about the factor.  Here they are:    

I'll give away one finding right away.  The most decisive factor, shown at the right of the lowermost column, is party affiliation. D's voted for Clinton. R's voted for Trump.

But after that, it's pretty clear that the close runner-up for most decisive factor in how people voted is the issue at the left of the middle column, which was ...

the economy!

That's right. The single most decisive factor in the 2016 vote was how people felt about the economy.

How can that be true?  After all, haven't we all heard that racism was dispositive?

The clue is in the difference in slope between the red and gray lines.  Even though the economy was the single most important issue in 2016, it was relatively *less* important than it was in 2012. Graphically it was less vertical in 2016 than 2012. And the difference disfavored democrats.  The economy was equally dispositive for conservatives, but less dispositive for liberals. Put another way, economic progressives voted less in lock-step on the economy in 2016 than they did in 2012.

And where did those votes go in 2016?  The three issues in the top column, and especially the issue at the top left.

But let's take them in reverse order.  The issue at the top right is nearly horizontal -- i.e., it made little difference to voters, but while it didn't motivate progressives at all, it did have a slight affect on conservative votes compared with 2012.  That issue is:

attitudes towards Muslims.

The issue at the top middle had some effect both in 2012, but moreso in 2016. It motivated both conservatives and progressives, but conservatives relatively more.  That issue is:

attitudes towards black people.  There more than anywhere is your racism, and while the issue did drive progressive voters to Clinton, it drove even more racist voters to Trump. 

   

This is an unfortunate dynamic for the left.  What it appears to mean is, both progressives and conservatives were motivated by the "Black Lives Matter" movement and protests against police violence. But while the protests may have convinced some progressives, they drove EVEN MORE CONSERVATIVES to harden, and act on, their racial views.  In other words, in electoral terms the protest movement was actually COUNTERproductive.

Finally, here is the social issue (top left among the nine) which most drove voter behavior in 2016 compared with 2012:

attitudes towards immigration. Note that, it too motivated both progressives and conservatives, and it too was more decisive in voting behavior on the right than on the left.

To get to the heart of the matter, it wasn't racism per se which drove the electoral college victory to Trump. It was Xenophobia, and anger in particular directed at illegal aliens. While probably about half of the Muslim immigrants to the US come from countries in the Middle East and are white, many also are from places like south Asia and are not white. Yet feelings about Muslims barely moved the needle.

Further, if it were racism per se which was most dispositive, we would expect to see a bigger correlation with feelings towards black people than feelings towards immigrants. Not only was that not the case, but the *reverse* was true.

In an article only a week after the election, Loren Collingwood wrote in the Washington Post that:

 I also found evidence consistent with the "racial threat" hypothesis. As shown by the orange dotted line in the graph, Trump’s vote was higher in counties where the number of Latinos has increased significantly since 2000. This suggests that some voters may have supported Trump as a way of expressing white identity in an increasingly diverse nation. 

... Trump also did better in counties experiencing a loss in manufacturing since 2000.

Here's her graph: 

It's worth noting that the unemployment rate among Latinos (I'm using men in the non-seasonal adjusted graph below) has declined further - from an admittedly higher peak -  in this expansion than that among whites:

At the height of the tech boom in 1999, male Latino unemployment was 1.3% higher than that of white males. In 2006 and again in June of this year, it was only 0.1% higher.

Since roughly 11 million, or 1/4 of the total Latino population, are illegals/undocumented workers, it is not difficult to imagine whites (and maybe some blacks) seething that illegals have taken some of their jobs, in a very long and still-incomplete recovery. Is that "economic anxiety" or racism or Xenophobia, or both?

 And by the way, a similar dynamic about the volume and nature of immigation, by the way, appears to be behind the big decline in the center left Social Democratic parties of Europe. 

Further, I wish I had kept the cite, but I recall reading stories of white working class voters who went to Trump rallies and participated in all of the chants, who supported him in part because they didn't take his racial rhetoric seriously.  They treated it as all part of the show. The distinction is worth emphasizing -- if I am a minority, I may not like a person who is willing to overlook that racial rhetoric, but it is different from a person who actually *agrees with* that racial rhetoric. More succinctly, if it comes down to whether you win or lose political power, you may be dead set against compromising with the latter, but what about the former?

I think there are three big lessons here for future elections: 

1. Even if you want to embrace the importance of social issues, you simply cannot ignore the economy,  which remains the single most important issue to most voters. As I have pointed out a number of times, econometric models did a very good job in 2016 forecasting a narrow popular vote advantage for the incumbent party.
2. Social issues should not be highlighted in ways which drive your supporters to the polls, but drive your opponents to the polls even more.
3. While I unequivocally support a path to legalization, and ultimately citizenship, for DREAMers, it is crystal clear from the experiences in both the US an d Europe that progressive parties have to come to grips with reasonable restrictions on immigration, and enforcement of, immigration laws.

Weekly Indicators for September 18 - 22 at XE.com

 - by New Deal democrat

My Weekly Indicators post is up at XE.com.

The biggest news was in purchase mortgage applications, hurricane adjusted jobless claims, and stock prices.

On August housing permits and starts, curb your enthusiasm

 - by New Deal democrat

Hurricanes Harvey and Irma together affected over 10% of the housing market at minimum. That's one of three good reasons to take the good permits number with a grain of salt.

This is a two part post I have up at XE.com.

Part 1 is here.
Part 2 is here.

Hurricane adjusted initial jobless claims for the week of September 9: 229,000

 - by New Deal democrat

I am repeating an exercise I undertook in 2012 when Superstorm Sandy disrupted the initial claims data: estimating what the initial jobless claims would have been, but for the hurricanes.

In 2012 I created that adjustment by backing out the affected states (NY and NJ) from the non-seasonally adjusted data.  That gave me the number of initial claims filed in the other 48 states.  I compared that with the same metric one year earlier, and multiplied by the seasonal adjustment.

That gave me the number if the affected states had the same relative number of claims during the given week, as all of the unaffected states.  In 2012, it showed that Sandy was not masking any underlying weakness in the economy.

The state by state data is released with a one week delay.  So what follows is the analysis for the week of September 9, the number for which was reported one week ago and revised this week to 282,000. Last week I found the adjusted number for September 2 was 239,000.  Last week I only had to back out Texas, but this week I have also backed out Florida.

Here is the table for the Week of September 10 in 2016 vs. September 9 this year:

Metric                              2016                   2017

Seasonally adjusted:       258,000              282,000

Adjustment for total:       1.33                   1.33

Not seasonally adjusted:  193,291             212,284 

Florida claims:                 7,493                 4,773 (!!! - yes, a decline this year)
Texas claims:                   13,432               52,024                

NSA claims ex-TX+FL    202,008             155,487

TX+FL as % of total:       10.8%                   n/a

2017 w/ TX+FL adjustment:  n/a              172,280

In both 2016 and 2017 the weekly seasonal adjustment was 1.33. Multiplying the non-seasonally adjusted total of 172,280, gives us the hurricane-adjusted initial jobless claims number for the week of September 9, 2017 of 229,000.

The underlying national trend in initial jobless claims remains very positive.

The asterisk in real median household income

 - by New Deal democrat

This is a follow-up to the post I wrote last week about the latest data on real median household income.

One of the things I notes is that "households" includes the millions that are composed of retirees, a burgeoning demographic due both to healthier longevities and the demographics of the Boomer generation.

This morning Jared Bernstein helpfully includes a graph of real median household income excluding those over age 65:

Households headed by working age adults did finally surpass their 2007 income, but were still 3.4% below the all-time highs of incomes of 2000.

But mainly I wanted to follow up on that break in the graph in 2013.  It was caused by a change in methodology by the Census Bureau.

Here's the graph I ran last week of real median household incomes at various quintiles and deciles:

So I was surprised a few days later to see another, more pessimistic graph floating around, purporting to show that only the top 20% of households had higher incomes than in 2000:

Note that this graph doesn't have any break. 

I traced the information back to its source, which turned out to be the Economic Policy Institute. And at the bottom of their article, I found a footnote explaining thus:

In other words, as best I can tell the E.P.I. simply took the 3.2% difference in the two methodologies in 2013 and projected it forward.

Now, let me state right up front that I don't know whether the data as represented by E.P.I. is correct or not.

But, neither does the E.P.I.

In fact, neither does anybody else, apparently including the Census Bureau itself.

That's because the Census Bureau hasn't provided data in any year since 2013 as to what the numbers would have been under their old methodology, so that we can form a basis of comparison.

By contrast, when the methodology for "real retail sales" was updated in the 1990s, the old data series continued to run for nearly 10 years, giving us an excellent basis for comparison, and confirming that the YoY changes were virtually identical, meaning that we can stitch the two series together confidently, giving us 70 years worth of data:

The Census Bureau didn't do that with real median household income, so there will *always* be a disconnect and a lack of ability to reliably and directly compare data from before 2013 with data afterward.  We'll always be guessing. 

This is a major problem for this important data series, and the Census Bureau should take steps, to the extent possible, to correct it.