Revisiting "The Mythology of Trump's Working Class Support"

[All Along The Watchtower]

Or, "Why Nate Silver's Political #Analysis is terrible, n = 5383276899":

This idea (which I was admittedly guilty of supporting) that Trump did not have substantial support among (white) working class voters seems to have really gotten a boost from this article by Silver.

You must be logged in to read this quote.

A few things.

1) You'll notice that the data here is for median household income. A married couple with husband and wife making $35k-$40k each = a household income between $70k and $80k. Doesn't seem that impressive, especially when you consider that

2) Republicans in general (and Republican primary voters,  I suspect, in particular) are not only more likely to be white and have higher-than-the-median incomes, but are also more likely to be married and (at least, compared to white liberals) have children - which are very much relevant when discussing one's means and financial responsibilities...

3) Republican primary voters are better off than Democratic primary voters, both Democratic and Republican primary voters are better off than general election-only voters, voters in general are better off than non-voters...

4) Measuring someone's social class by their household income at a given point in time seems very...um, simplistic, to be charitable. Especially when we're talking about a very, very broad group of people in terms of income distribution, and with a lot of those people in the "vast middle" really not doing so well these days...

5) Compare Trump's voters to say, Rubio's voters, Kasich's voters, or even Cruz's voters. And compare the non-Trump Republican candidates' numbers among the more modestly educated (i.e. those who don't possess a four-year degree) with their numbers among the higher cohorts. I don't think I need to spell any more of this out.

tl; dr Trump did indeed, have (and has) an unusually (although not uniformly - something that we also must be mindful about) strong appeal among working class white voters for a Republican candidate, in spite of the fact that - like all Republican candidates - his voters still skewed toward the white and the affluent.

PS: F**k Nate Silver.

[Mr. Morden]

tl; dr Trump did indeed, have (and has) an unusually (although not uniformly - something that we also must be mindful about) strong appeal among working class white voters for a Republican candidate, in spite of the fact that - like all Republican candidates - his voters still skewed toward the white and the affluent.

I'm confused, because I thought that's what the article itself said.  So I'm unclear on where the disagreement is.  It sounds like it's a difference in emphasis--a difference in where you put the "but"....

Silver: Yes, Trump's voters are poorer than the voters for other Republican candidates, *but* they're still wealthier than the average American.

You: Yes, Trump's voters are wealthier than the average American, *but* they're not as wealthy as those of other Republican candidates.

Where's the disagreement between those two propositions?

[hopper]

The difference was non-college whites(+42% in popular vote margin vs HRC) as opposed to College Educated Whites(+2% popular vote margin vs HRC) it wasn't really income based.

[Xing]

The difference was non-college whites(+42% in popular vote margin vs HRC) as opposed to College Educated Whites(+2% popular vote margin vs HRC) it wasn't really income based.

This is the key. Education was a much better indicator of Trump's level of support than income.

[NOVA Green]

The difference was non-college whites(+42% in popular vote margin vs HRC) as opposed to College Educated Whites(+2% popular vote margin vs HRC) it wasn't really income based.

This is the key. Education was a much better indicator of Trump's level of support than income.

What is completely neglected here in this discussion is the correlation between age & educational levels.

Older Americans (65+ in particular) are much more likely to be White and significantly lower levels of educational attainment than the general population.

Nothing wrong with that at all.... for my parent's generation (Baby Boomers) going to college/university was a luxury. and for both my parents they were the first in their families to have a four year degree.

Back in those days, many Americans believed in the American Dream, where you could secure your future and your family's future getting a decent paying union job working the line, or even in a grocery store, at a time where a thing called "pattern bargaining" still existed, before it was gutted under Ronald Reagan and the shift to Plant-by-Plant Union contracts.

If you were Male, you had a significant chance of getting shipped over to 'Nam, if your lottery number came up, regardless of educational attainment or background, like some of my older friends and almost happened to my Father, except he was rejected from service because of medical conditions.

It is only relatively recently where a four year degree has become a virtual necessity to "succeed" in this fantasy called the American Dream, and without a two year vocational degree one is lucky to be able to land a decent paying first job outside of a gas station or retail store.

Although I agree with the premise that income "Trumped education", it's really easy to sit here from a mile high view and gaze down without looking at the yuuge generational gap in the '16 Presidential Election, that in many ways had a stronger correlation than education/income gaps among White voters .

I do not and will not subscribe to the theory that some of the Liberals and Conservatives toss out on this Forum nor elsewhere, that somehow WWC voters are " racists, lazy, stupid, ignorant, voting against their own economic interests", and all of the other crap that pops up on a regular basis like some type of virulent STD or bedbugs.

Sure, Non-College Whites swung heavily towards Trump in '16 no debate and no question on that fact.

Many of the Non-College Whites (Less than 4 yr degree?) are those 55+, and that number increases dramatically each year you add on....

Trump performed extremely well, precisely because he was able to max out the White vote, particularly among Seniors, and also Gen X-ers, many of whom voted for Gore/Kerry/Obama between 2000 and 2012.

Not rocket science here... sure there were many parts of the country where WWC Voters in ancestral Democratic areas went 3rd Party, but where the votes actually mattered in Blue Collar PA cities, Youngstown Ohio, Macomb County Michigan, factory towns in Eastern Iowa, Racine and Kenosha Wisconsin, it's pretty clear that there were a ton of Obama '08/'12 voters that went Trump in '16.

[Mr. Morden]

Although I agree with the premise that income "Trumped education", it's really easy to sit here from a mile high view and gaze down without looking at the yuuge generational gap in the '16 Presidential Election, that in many ways had a stronger correlation than education/income gaps among White voters .

OK, but the generation gap was larger in 2012 than it was in 2016.

From the exit polls...

Obama vs. Romney:
age 18-29: Obama +23
age 65+: Romney +12
college graduates: Obama +2
non-college graduates: Obama +4

Clinton vs. Trump:
age 18-29: Clinton +19
age 65+: Trump +7
college graduates: Clinton +10
non-college graduates: Trump +7

So compared to Romney voters, Trump voters were both younger and less educated.

[hopper]

The difference was non-college whites(+42% in popular vote margin vs HRC) as opposed to College Educated Whites(+2% popular vote margin vs HRC) it wasn't really income based.

This is the key. Education was a much better indicator of Trump's level of support than income.

What is completely neglected here in this discussion is the correlation between age & educational levels.

Older Americans (65+ in particular) are much more likely to be White and significantly lower levels of educational attainment than the general population.

Nothing wrong with that at all.... for my parent's generation (Baby Boomers) going to college/university was a luxury. and for both my parents they were the first in their families to have a four year degree.

Back in those days, many Americans believed in the American Dream, where you could secure your future and your family's future getting a decent paying union job working the line, or even in a grocery store, at a time where a thing called "pattern bargaining" still existed, before it was gutted under Ronald Reagan and the shift to Plant-by-Plant Union contracts.

If you were Male, you had a significant chance of getting shipped over to 'Nam, if your lottery number came up, regardless of educational attainment or background, like some of my older friends and almost happened to my Father, except he was rejected from service because of medical conditions.

It is only relatively recently where a four year degree has become a virtual necessity to "succeed" in this fantasy called the American Dream, and without a two year vocational degree one is lucky to be able to land a decent paying first job outside of a gas station or retail store.

Although I agree with the premise that income "Trumped education", it's really easy to sit here from a mile high view and gaze down without looking at the yuuge generational gap in the '16 Presidential Election, that in many ways had a stronger correlation than education/income gaps among White voters .

I do not and will not subscribe to the theory that some of the Liberals and Conservatives toss out on this Forum nor elsewhere, that somehow WWC voters are " racists, lazy, stupid, ignorant, voting against their own economic interests", and all of the other crap that pops up on a regular basis like some type of virulent STD or bedbugs.

Sure, Non-College Whites swung heavily towards Trump in '16 no debate and no question on that fact.

Many of the Non-College Whites (Less than 4 yr degree?) are those 55+, and that number increases dramatically each year you add on....

Trump performed extremely well, precisely because he was able to max out the White vote, particularly among Seniors, and also Gen X-ers, many of whom voted for Gore/Kerry/Obama between 2000 and 2012.

Not rocket science here... sure there were many parts of the country where WWC Voters in ancestral Democratic areas went 3rd Party, but where the votes actually mattered in Blue Collar PA cities, Youngstown Ohio, Macomb County Michigan, factory towns in Eastern Iowa, Racine and Kenosha Wisconsin, it's pretty clear that there were a ton of Obama '08/'12 voters that went Trump in '16.

No the cutoff is age 45+ not age 65+ as far as Whites go. Here is the age white demographic breakdown whites from Larry Sabato's Website off of a recent article how "How Midterms are different from Presidential Elections." I'm not sure if the demographic breakdown from the article is VAP or just from the census though. The average White% of the electorate was 70% from the 2016 exit polls I know that.

Ages 18-29: 56% White
Ages 30-44: 58% White
Ages 45-64: 70% White
Ages 65%: 79%

[Kringla Heimsins]

The difference was non-college whites(+42% in popular vote margin vs HRC) as opposed to College Educated Whites(+2% popular vote margin vs HRC) it wasn't really income based.

[RFayette]

Where are these wealthy Trump voters coming from anyways?  He pretty much lost the vast majority of wealthy counties in this country (often overwhelmingly).

Despite having lost ground with them, Trump still did very well in places like Shelby County, AL, Williamson County, TN, etc.  He also did well with high income voters without a college degree, which are more likely to be found in poorer counties.

[Bismarck]

Where are these wealthy Trump voters coming from anyways?  He pretty much lost the vast majority of wealthy counties in this country (often overwhelmingly).

Despite having lost ground with them, Trump still did very well in places like Shelby County, AL, Williamson County, TN, etc.  He also did well with high income voters without a college degree, which are more likely to be found in poorer counties.

Basically this. Hamilton County Indiana despite its strong D trend still gave Trump  almost 60% of the vote. He also won the wealthiest county in New York, New Jersey, Colorado, Wisconsin, Ohio, Florida, (and I think most states although I haven't checked on that)

[NOVA Green]

Where are these wealthy Trump voters coming from anyways?  He pretty much lost the vast majority of wealthy counties in this country (often overwhelmingly).

Despite having lost ground with them, Trump still did very well in places like Shelby County, AL, Williamson County, TN, etc.  He also did well with high income voters without a college degree, which are more likely to be found in poorer counties.

Fremont hits the nail on the head here....

Not sure about the claim that "Trump lost the vast majority of wealthy counties".

Wealth is relative, depending upon the particular housing expenses in various regional Metro areas.

It does appear that there was a huge swing against Trump, and to a lesser extent towards Clinton in many wealthy areas throughout the United States.

That being said, Fremont is correct that even in wealthy areas in places in Texas, Tennessee, etc.... that wealthy voters did not back Clinton among wealthy White voters in most states of the old confederacy....

We have already discussed elsewhere and I ran a few numbers in wealthy parts of NoVA, but the claim that Trump lost of the vote of those families with a >$250k/Yr MHI throughout the United States has yet to be proven....

There is a whole entire thread about the wealthiest places in America and how they voted in '16, and looking forward to your additional contributions over there), but conflating "wealthiest counties" with "wealthiest places" without controlling for variables such as the cost of housing and such, is a bit spurious at best, considering that generally the higher income counties are also in extremely expensive metro areas, where much of the additional income does not correlate with home ownership, let alone take home income once the rent/mortgage is paid.

Unfortunately, it is extremely difficult to isolate wealthy voters in "rural" America outside of Metro areas, but again Fremont has a strong point here....

Link to thread on wealthy places in US voting thread.... jump in and contribute! We need a ton more data on this topic....

https://uselectionatlas.org/FORUM/index.php?topic=259050.0

[NOVA Green]

Where are these wealthy Trump voters coming from anyways?  He pretty much lost the vast majority of wealthy counties in this country (often overwhelmingly).

Despite having lost ground with them, Trump still did very well in places like Shelby County, AL, Williamson County, TN, etc.  He also did well with high income voters without a college degree, which are more likely to be found in poorer counties.

Basically this. Hamilton County Indiana despite its strong D trend still gave Trump  almost 60% of the vote. He also won the wealthiest county in New York, New Jersey, Colorado, Wisconsin, Ohio, Florida, (and I think most states although I haven't checked on that)

Thanks St Alphonso!

Care to pull the wealthiest counties in these states, to save myself some time to see what the data shows for the wealthiest places within those states, assuming precinct/city/township level data is available....

[RI]

Where are these wealthy Trump voters coming from anyways?  He pretty much lost the vast majority of wealthy counties in this country (often overwhelmingly).

Despite having lost ground with them, Trump still did very well in places like Shelby County, AL, Williamson County, TN, etc.  He also did well with high income voters without a college degree, which are more likely to be found in poorer counties.

Basically this. Hamilton County Indiana despite its strong D trend still gave Trump  almost 60% of the vote. He also won the wealthiest county in New York, New Jersey, Colorado, Wisconsin, Ohio, Florida, (and I think most states although I haven't checked on that)

Thanks St Alphonso!

Care to pull the wealthiest counties in these states, to save myself some time to see what the data shows for the wealthiest places within those states, assuming precinct/city/township level data is available....

Among precincts situated in a census tract with $x in median household income, Trump averaged y percent of the vote:

Median HH Inc.	Trump %	Swing to Clinton
>$100,000	38.01%	11.44%
>$120,000	36.07%	15.33%
>$140,000	34.41%	18.57%
>$160,000	32.35%	20.92%
>$180,000	30.50%	22.77%
>$200,000	30.77%	24.32%
>$220,000	29.85%	28.07%

Now, due to the ecological fallacy, we can't say for sure that there's a negative individual level relationship between Trump vote and income, but there certainly is one at the precinct level.