Well I hope the Dems are proud of themselves

[Taran Wanderer]

One of the big problems in our best universities is that they’re not getting bigger (or multiplying).  So as the population grows, as the population they wish to draw from expands across continents, and as our increasingly complex world has arguably more need for the people they can teach, these institutions are becoming more and more exclusive.  Meanwhile, the best ones are growing bigger and bigger endowments which supercharge their capability, making them even more desirable, and again, more exclusive.

In my state, we’ve seen a new state university spawned over the past 10-15 years.  It started as an outgrowth of another institution, but I expect it will eventually be independent.  I would like to see more of this.  I would love to see Harvard and Stanford the other top private institutions in the country, which are largely located on the coasts, found new universities in less economically advantaged parts of the country.

This could be an economic engine for those areas, but I don’t think the benefits would only flow in one direction.  I think being closer to the challenges in these areas, and being given the opportunity to listen to the challenges and be part of solving them, would foster great dialogue and understanding.  It would also open more seats for potential students and spread the wealth of these institutions’ endowments rather than focusing it in areas that are already wealthy.

Maybe this is off topic for this thread, but just something I think about.

[Kris]

One of the big problems in our best universities is that they’re not getting bigger (or multiplying).  So as the population grows, as the population they wish to draw from expands across continents, and as our increasingly complex world has arguably more need for the people they can teach, these institutions are becoming more and more exclusive.  Meanwhile, the best ones are growing bigger and bigger endowments which supercharge their capability, making them even more desirable, and again, more exclusive.

In my state, we’ve seen a new state university spawned over the past 10-15 years.  It started as an outgrowth of another institution, but I expect it will eventually be independent.  I would like to see more of this.  I would love to see Harvard and Stanford the other top private institutions in the country, which are largely located on the coasts, found new universities in less economically advantaged parts of the country.

This could be an economic engine for those areas, but I don’t think the benefits would only flow in one direction.  I think being closer to the challenges in these areas, and being given the opportunity to listen to the challenges and be part of solving them, would foster great dialogue and understanding.  It would also open more seats for potential students and spread the wealth of these institutions’ endowments rather than focusing it in areas that are already wealthy.

Maybe this is off topic for this thread, but just something I think about.

In the US, college enrollments have been falling for a while now. It’s spoken of as a crisis in higher ed circles.

There are many possible intersecting reasons for this, but one highly likely reason is the astronomical cost of a university education these days. Going into debt for the rest of your life with no possibility of ever discharging the loan in bankruptcy is not a fantastically attractive prospect, as it turns out.

[twinstudy]

Do state colleges and community colleges charge lots? I had always understood them to be relatively affordable.

Plenty of elite schools waive or reduce tuition for certain students based on academic ability or family income. Half of Harvard undergrads and two-thirds of Stanford undergrads receive some form of financial aid. One-quarter of Harvard undergrads pay nothing at all after grants.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

[nereo]

Do state colleges and community colleges charge lots? I had always understood them to be relatively affordable.

Plenty of elite schools waive or reduce tuition for certain students based on academic ability or family income. Half of Harvard undergrads and two-thirds of Stanford undergrads receive some form of financial aid. One-quarter of Harvard undergrads pay nothing at all after grants.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

It varies wildly from one public university to the next. Ours estimates the cost at $27k/yr for everything, and we are towards the low end. Other public universities can exceed $40k. Here I’m including tuition, fees, recommended supplies and rent.

Yes,there are several (I wouldn’t say “plenty” ) of elite schools the waive tuition for students from lower income families.  But each year about 4 Million students graduate high school, and Harvard admits about 1,600.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

Ok… but why?  You’ve already said that the very biggest names don’t charge tuition for lower income students. There are other models, such as our state university systems that (until very recently) offered some of the lowest cost and most recognized names (Berkeley, Michigan, UVA etc). In the UK they have capped tuition at all higher education which includes Cambridge and St Andrews.

It gets us into the discussion about whether college/university should be financially acceptable to all. Until q the early 20th century high school was not universally available, and was frequently only accessible to the wealthy.

[Villanelle]

Do state colleges and community colleges charge lots? I had always understood them to be relatively affordable.

Plenty of elite schools waive or reduce tuition for certain students based on academic ability or family income. Half of Harvard undergrads and two-thirds of Stanford undergrads receive some form of financial aid. One-quarter of Harvard undergrads pay nothing at all after grants.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

UC berkely is a state school.  A quick google tells me 4 years will cost about $184,032 for in-state students.  I think that's before books, housing, etc. 

I also found a stat (with no year posted so I don't know how recent) that a four-year program at a public university in the U.S. averages $111,760.  It doesn't specify whether this is for in-state residents or an average of what all students pay.

[Sandi_k]

Do state colleges and community colleges charge lots? I had always understood them to be relatively affordable.

Plenty of elite schools waive or reduce tuition for certain students based on academic ability or family income. Half of Harvard undergrads and two-thirds of Stanford undergrads receive some form of financial aid. One-quarter of Harvard undergrads pay nothing at all after grants.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

UC berkely is a state school.  A quick google tells me 4 years will cost about $184,032 for in-state students.  I think that's before books, housing, etc. 

I work for UC. For a CA resident undergraduate, tuition and fees and medical coverage, it's $10,500 per semester, or $21k per year. That does not include housing.

https://registrar.berkeley.edu/tuition-fees-residency/tuition-fees/fee-schedule/

If you are a non-resident of CA, the cost is $27,650 per semester, or $55,300 per year, not including housing.

[wenchsenior]

One of the big problems in our best universities is that they’re not getting bigger (or multiplying).  So as the population grows, as the population they wish to draw from expands across continents, and as our increasingly complex world has arguably more need for the people they can teach, these institutions are becoming more and more exclusive.  Meanwhile, the best ones are growing bigger and bigger endowments which supercharge their capability, making them even more desirable, and again, more exclusive.

In my state, we’ve seen a new state university spawned over the past 10-15 years.  It started as an outgrowth of another institution, but I expect it will eventually be independent.  I would like to see more of this.  I would love to see Harvard and Stanford the other top private institutions in the country, which are largely located on the coasts, found new universities in less economically advantaged parts of the country.

This could be an economic engine for those areas, but I don’t think the benefits would only flow in one direction.  I think being closer to the challenges in these areas, and being given the opportunity to listen to the challenges and be part of solving them, would foster great dialogue and understanding.  It would also open more seats for potential students and spread the wealth of these institutions’ endowments rather than focusing it in areas that are already wealthy.

Maybe this is off topic for this thread, but just something I think about.

In the US, college enrollments have been falling for a while now. It’s spoken of as a crisis in higher ed circles.

There are many possible intersecting reasons for this, but one highly likely reason is the astronomical cost of a university education these days. Going into debt for the rest of your life with no possibility of ever discharging the loan in bankruptcy is not a fantastically attractive prospect, as it turns out.

That's interesting. Is it total raw enrollment, or enrollment as a percentage of total U.S. population? B/c at our state university (large) enrollment has been climbing in general for a long time, and most recently we've seen 5 years of consecutive growth in enrollment.

[Kris]

One of the big problems in our best universities is that they’re not getting bigger (or multiplying).  So as the population grows, as the population they wish to draw from expands across continents, and as our increasingly complex world has arguably more need for the people they can teach, these institutions are becoming more and more exclusive.  Meanwhile, the best ones are growing bigger and bigger endowments which supercharge their capability, making them even more desirable, and again, more exclusive.

In my state, we’ve seen a new state university spawned over the past 10-15 years.  It started as an outgrowth of another institution, but I expect it will eventually be independent.  I would like to see more of this.  I would love to see Harvard and Stanford the other top private institutions in the country, which are largely located on the coasts, found new universities in less economically advantaged parts of the country.

This could be an economic engine for those areas, but I don’t think the benefits would only flow in one direction.  I think being closer to the challenges in these areas, and being given the opportunity to listen to the challenges and be part of solving them, would foster great dialogue and understanding.  It would also open more seats for potential students and spread the wealth of these institutions’ endowments rather than focusing it in areas that are already wealthy.

Maybe this is off topic for this thread, but just something I think about.

In the US, college enrollments have been falling for a while now. It’s spoken of as a crisis in higher ed circles.

There are many possible intersecting reasons for this, but one highly likely reason is the astronomical cost of a university education these days. Going into debt for the rest of your life with no possibility of ever discharging the loan in bankruptcy is not a fantastically attractive prospect, as it turns out.

That's interesting. Is it total raw enrollment, or enrollment as a percentage of total U.S. population? B/c at our state university (large) enrollment has been climbing in general for a long time, and most recently we've seen 5 years of consecutive growth in enrollment.

That’s overall. College enrollment peaked in 2010.

https://educationdata.org/college-enrollment-statistics

[lhamo]

It's largely demographic -- US birth rates have dropped dramatically since 2008, which means the number of graduating high schoolers has peaked.

https://www.bestcolleges.com/news/analysis/looming-enrollment-cliff-poses-serious-threat-to-colleges/

It is more of an issue for small/expensive schools that already struggle to fill their seats.  High prestige schools with sub 20% enrollment rates won't be affected much.  Good quality state schools that are more affordable will probably also be fine.  States with good population growth will also be fine.

[rocketpj]

They could, of course, stop spending so much money on sports and adminstrative bloat, lower tuitions and increase academics, and become more attractive places to learn.  But current US college systems see that as insanity.

[reeshau]

They could, of course, stop spending so much money on sports and adminstrative bloat, lower tuitions and increase academics, and become more attractive places to learn.  But current US college systems see that as insanity.

Don't take away the lazy rivers!

https://www.forbes.com/sites/michaelhorn/2019/11/14/why-lazy-rivers-have-their-place-on-college-campusesand-yet-still-might-just-be-lazy/

Think of the lost opportunity of "living like a college student."  No need for hedonic adaptation--you're already screwed!

[Kris]

They could, of course, stop spending so much money on sports and adminstrative bloat, lower tuitions and increase academics, and become more attractive places to learn.  But current US college systems see that as insanity.

Don't take away the lazy rivers!

https://www.forbes.com/sites/michaelhorn/2019/11/14/why-lazy-rivers-have-their-place-on-college-campusesand-yet-still-might-just-be-lazy/

Think of the lost opportunity of "living like a college student."  No need for hedonic adaptation--you're already screwed!

Yeah, this is one of the sadder aspects. In a time of dwindling public support for higher ed, colleges and universities started to compete fiercely with other universities for students' tuition dollars, which has meant trying to make their campuses as attractive and fun-sounding as possible. Which costs a lot of money. Vicious cycle.

[wenchsenior]

They could, of course, stop spending so much money on sports and adminstrative bloat, lower tuitions and increase academics, and become more attractive places to learn.  But current US college systems see that as insanity.

Don't take away the lazy rivers!

https://www.forbes.com/sites/michaelhorn/2019/11/14/why-lazy-rivers-have-their-place-on-college-campusesand-yet-still-might-just-be-lazy/



Think of the lost opportunity of "living like a college student."  No need for hedonic adaptation--you're already screwed!

Have to admit... I ADORE our insane over the top aquatic facility, including the lazy river. Heading down there in about 15 minutes in fact.

[Taran Wanderer]

They could, of course, stop spending so much money on sports and adminstrative bloat, lower tuitions and increase academics, and become more attractive places to learn.  But current US college systems see that as insanity.

I love this idea, but then they wouldn’t have a marketing staff to market a boring focused education primarily dedicated to actual learning. Might be hard to attract students.

And while I agree with the assessment (administrative bloat, etc.), this problem isn’t limited to college and university. It’s everywhere. K-13 education, healthcare, corporations, and broader government. I don’t have solutions, but I do like to complain about it.

[twinstudy]

Do state colleges and community colleges charge lots? I had always understood them to be relatively affordable.

Plenty of elite schools waive or reduce tuition for certain students based on academic ability or family income. Half of Harvard undergrads and two-thirds of Stanford undergrads receive some form of financial aid. One-quarter of Harvard undergrads pay nothing at all after grants.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

It varies wildly from one public university to the next. Ours estimates the cost at $27k/yr for everything, and we are towards the low end. Other public universities can exceed $40k. Here I’m including tuition, fees, recommended supplies and rent.

Yes,there are several (I wouldn’t say “plenty” ) of elite schools the waive tuition for students from lower income families.  But each year about 4 Million students graduate high school, and Harvard admits about 1,600.

If you are going to go for a big name degree and you do not have top marks then you should be prepared to fork out. It is supply and demand.

Ok… but why?  You’ve already said that the very biggest names don’t charge tuition for lower income students. There are other models, such as our state university systems that (until very recently) offered some of the lowest cost and most recognized names (Berkeley, Michigan, UVA etc). In the UK they have capped tuition at all higher education which includes Cambridge and St Andrews.

It gets us into the discussion about whether college/university should be financially acceptable to all. Until q the early 20th century high school was not universally available, and was frequently only accessible to the wealthy.

Including supplies and rent is going to push the fees up. $27k a year sounds very reasonable to me. University is meant to be an investment after all. The returns should easily exceed the $27k a year cost, unless you're studying basket weaving or something.

University should be within the reach of all high-achieving students - this means that those from low-income families should get subsidised/waived fees. But this is contingent on getting good marks. No one should expect to have a free ride to uni without the academic merit to back it up. Australia does subsidised Commonwealth places for those with higher marks and full-fee places for those with lower marks - I think this is a fair system.

[rocketpj]

Pretty much every other country in the world is very aware that an educated population is a priceless asset, and doesn't seek to immiserate students by charging insanely high tuition and piling nonrich students with insurmoutable debtloads.

I love this idea, but then they wouldn’t have a marketing staff to market a boring focused education primarily dedicated to actual learning. Might be hard to attract students.

Charge 20% of the typical tuition, provide access to decent housing and provide a high quality education.  It isn't about marketing, it isn't supposed to be a business.  You aren't selling mattresses.  Offer a high quality education at a fraction of the cost and students will flock to the school.

[Taran Wanderer]

Sorry, I was being sarcastic.

This is another example of a systemic flaw in our accounting systems.  We measure the cost to produce something - widgets, students, healthcare, etc. and report it as COGS - Cost Of Goods Sold.  As we all know, cost is something to me minimized, so we work to minimize COGS both on a unit basis and an overall basis.  This helps pump up Gross Margin.  Then, costs of administration, accounting, sales, marketing, executives, and other overhead costs can expand to consume that Gross Margin, leaving whatever is an acceptable Net Margin.  I have to wonder how our society would be different if instead of working to minimize COGS, we worked to maximize quality (which might take higher COGS) and minimize all the overhead.  More people would be focused on the doing instead of the management of the doing.  The “doing” (making, teaching, nursing, building, soldiering, farming, etc.) could be higher valued and esteemed, and the accounting, marketing, selling, and other inflated bureaucracy jobs could be less necessary and less valued.

[twinstudy]

Pretty much every other country in the world is very aware that an educated population is a priceless asset, and doesn't seek to immiserate students by charging insanely high tuition and piling nonrich students with insurmoutable debtloads.

I love this idea, but then they wouldn’t have a marketing staff to market a boring focused education primarily dedicated to actual learning. Might be hard to attract students.

Charge 20% of the typical tuition, provide access to decent housing and provide a high quality education.  It isn't about marketing, it isn't supposed to be a business.  You aren't selling mattresses.  Offer a high quality education at a fraction of the cost and students will flock to the school.

Whether you like it or not, education is a positional good. If university became accessible (financially, academically  or both) to 80% of the population it would no longer have any signifying power and would instead be equivalent to a high school diploma. Sought-after degrees like finance, law and medicine are only sought after because they have high entry requirements and signify something about the holder - hopefully, that the holder was able to blitz the SATs, maintain a very good GPA throughout high school, get into a demanding course and excel.

So whatever the situation is, an elite degree will always be expensive in one way or another - either financially, or academically (i.e. only the very best students can get it).

Sure, community colleges should offer plain, low cost degrees and associate degrees. Which would be sufficient for entry level jobs and basic skilled jobs. I had a look at this page and it looks like a lot of community colleges offer in-state tuition for $5k a year which seems more than reasonable to me.

https://educationdata.org/average-cost-of-community-college

[Ron Scott]

LOL—“I LOVE THE EDUCATED.”

Yet we digress…

So we ditched Biden for Harris, picked a couple VP candidates who are vying for the yuck yuck local-yokel award, got over the honeymoons from both parties’ glam-rock shows, had THE BIG DEBATE, a couple assassination attempts, and the inflation horror seems to be in remission. Cool.

I’m still wondering if the Ds got this. Polls hot out of the oven seem to be leaning Harris, but not by much. And we’ve still got about 7 weeks left.

You Dems feeling good?? Proud?

[Kris]

LOL—“I LOVE THE EDUCATED.”

Yet we digress…

So we ditched Biden for Harris, picked a couple VP candidates who are vying for the yuck yuck local-yokel award, got over the honeymoons from both parties’ glam-rock shows, had THE BIG DEBATE, a couple assassination attempts, and the inflation horror seems to be in remission. Cool.

I’m still wondering if the Ds got this. Polls hot out of the oven seem to be leaning Harris, but not by much. And we’ve still got about 7 weeks left.

You Dems feeling good?? Proud?

Condescending tone aside, I'm feeling hopeful.

[dividendman]

LOL—“I LOVE THE EDUCATED.”

Yet we digress…

So we ditched Biden for Harris, picked a couple VP candidates who are vying for the yuck yuck local-yokel award, got over the honeymoons from both parties’ glam-rock shows, had THE BIG DEBATE, a couple assassination attempts, and the inflation horror seems to be in remission. Cool.

I’m still wondering if the Ds got this. Polls hot out of the oven seem to be leaning Harris, but not by much. And we’ve still got about 7 weeks left.

You Dems feeling good?? Proud?

All swing states are in the margin of error of the polls... it's 50/50.

[NorCal]

LOL—“I LOVE THE EDUCATED.”

Yet we digress…

So we ditched Biden for Harris, picked a couple VP candidates who are vying for the yuck yuck local-yokel award, got over the honeymoons from both parties’ glam-rock shows, had THE BIG DEBATE, a couple assassination attempts, and the inflation horror seems to be in remission. Cool.

I’m still wondering if the Ds got this. Polls hot out of the oven seem to be leaning Harris, but not by much. And we’ve still got about 7 weeks left.

You Dems feeling good?? Proud?

All swing states are in the margin of error of the polls... it's 50/50.

As someone who's job it is to misuse statistics on a regular basis, I think it's more likely than not going to be Kamala.  It's not a sure thing, but I think it's more likely.

My general observation on polling is that the polls are really good at identifying which demographic groups are going to go for each candidate, and by what margin.  I'm sure the polls know exactly how 38-42 year old female Ukelele players are going to vote this time around.

What the polls have trouble getting right is voter turnout.  The candidates aren't competing with each other.  They're competing with the couch.  And I don't mean JD Vance.

Trump had a statistically improbably victory in 2016 because traditional democrat supporters didn't show up in the numbers expected.  And the diehard MAGA demographics voted at an unprecedented rate in 2016.

The real question is where are the polls overestimating or underestimating turnout assumptions?  Do I believe that the polls are underestimating Trump's turnout numbers like they did in 2016?  Or do I believe a larger percent of Kamala supporters will show up than expected?

I suspect there's a number of people in the middle that just won't show up for Trump.  Maybe they stay home, leave the presidential box blank, or vote for a third party candidate.  I think Kamala will drive turnout in a way Biden couldn't have.  It's a question of whether maybe 0.25% of potential Trump voters stay home and 0.25% incremental voters show up for Kamala in swing states.

[MustacheAndaHalf]

Trump had a statistically improbably victory in 2016 because traditional democrat supporters didn't show up in the numbers expected.  And the diehard MAGA demographics voted at an unprecedented rate in 2016.

Those same pollsters got it wrong again, in 2020 (Freudian slip: 2024).  They have never gotten Trump's poll numbers right, so I am suspicious of them doing so now.  A +1% Harris lead in swing states could easy fall to Trump for that reason.

Between Donald "I didn't lose 2020" Trump and Joe "What year is it?" Biden, I disliked both choices.  Harris is an improvement.

On a slightly related note, the Federal Reserve just made a 0.50% rate cut, rather than a 0.25% cut.  As the economy slows, the Fed is lowering rates to encourage economic activity (and also to get back to a normal level).  The larger cut will help the economy, which could help Harris.  Expect Trump to complain about it.

[Ron Scott]

On a slightly related note, the Federal Reserve just made a 0.50% rate cut, rather than a 0.25% cut.  As the economy slows, the Fed is lowering rates to encourage economic activity (and also to get back to a normal level).  The larger cut will help the economy, which could help Harris.  Expect Trump to complain about it.

My typical comment regarding the effectiveness of US institutions is that if it weren’t for the DoD and American businesses, we’d be a third-rate country. I was wrong not to include the Federal Reserve as they deserve a spot on the list.

The problem I have with the US economy—which is not the Fed’s fault—is with a lackluster growth in productivity. Productivity gains are magic, enabling more money to circulate in the economy without a concomitant increase in inflation…unless they’re low, which is where we’re at. I’d love the politicians to focus on this, but I’m not holding my breath.

[JLee]

On a slightly related note, the Federal Reserve just made a 0.50% rate cut, rather than a 0.25% cut.  As the economy slows, the Fed is lowering rates to encourage economic activity (and also to get back to a normal level).  The larger cut will help the economy, which could help Harris.  Expect Trump to complain about it.

My typical comment regarding the effectiveness of US institutions is that if it weren’t for the DoD and American businesses, we’d be a third-rate country. I was wrong not to include the Federal Reserve as they deserve a spot on the list.

The problem I have with the US economy—which is not the Fed’s fault—is with a lackluster growth in productivity. Productivity gains are magic, enabling more money to circulate in the economy without a concomitant increase in inflation…unless they’re low, which is where we’re at. I’d love the politicians to focus on this, but I’m not holding my breath.

https://www.economicstrategygroup.org/publication/in-brief-us-labor-productivity/

After decades of slow productivity growth, strong estimates over the past year suggest that period has come to an end.

Economist Robert Gordon has documented that the United States enjoyed a “special century” of productivity growth from just around the Civil War until the 1970s. Innovations during that time—including electricity, railroads, and the automobile—dramatically raised the productive capacity of US businesses. During the tail end of that century, from 1950 to 1970, worker productivity was growing at an average annual rate of 2.7 percent. Since then, the economy has experienced three distinct phases of productivity growth: a slowdown in the quarter-century from 1970 to 1994, when growth averaged 1.8 percent; a decade of faster growth from 1994 to 2004, when the internet revolution sped business productivity growth to a 2.9 percent rate; and finally a return to slower growth of 1.5 percent from 2004 to 2022. In the four quarters of 2023, however, productivity growth has experienced a renewed surge, averaging 2.7 percent—equal or similar to those prior waves of productivity growth.

[Ron Scott]

^^^  One year does not a trend make, but we shall see. Also, hoping MS and Amazon get the AI ball rolling on boosting value for businesses. Im dizzy with the hype, disappointed in the results to date.

[NorCal]

Trump had a statistically improbably victory in 2016 because traditional democrat supporters didn't show up in the numbers expected.  And the diehard MAGA demographics voted at an unprecedented rate in 2016.

Those same pollsters got it wrong again, in 2020 (Freudian slip: 2024).  They have never gotten Trump's poll numbers right, so I am suspicious of them doing so now.  A +1% Harris lead in swing states could easy fall to Trump for that reason.

Between Donald "I didn't lose 2020" Trump and Joe "What year is it?" Biden, I disliked both choices.  Harris is an improvement.

On a slightly related note, the Federal Reserve just made a 0.50% rate cut, rather than a 0.25% cut.  As the economy slows, the Fed is lowering rates to encourage economic activity (and also to get back to a normal level).  The larger cut will help the economy, which could help Harris.  Expect Trump to complain about it.

I agree that it's possible Trump's support is understated again.  This is where my prediction might fall short.  I generally recall that the pollsters did a better job in 2020 than 2016, but I've honestly forgotten the details.  But these pollsters are always correcting their models based on more recent elections.  So I don't think this will be a major factor, but we need an election to find out. 

If I have to handicap it, I'm still going to say Trump voters are probably marginally less likely to show up at the polls than a statistical model predicts and Harris voters marginally more likely.

I also wonder how the down-ballot races are going to play out.  While I don't have any specific predictions, I think there will be some surprises based on turnout in various counties/states.  I think a few legislators in "safe seats" may be packing up their bags.  I suspect part of the surprise-factor on voter turnout will vary heavily by geography.  Maybe an abortion amendment will make more people show up in some states.  Maybe the people in Springfield OH vote in overwhelming numbers because of the national attention (on both sides).  Down-ballot issues will play a big role in how motivated voters are. 

[MustacheAndaHalf]

Trump had a statistically improbably victory in 2016 because traditional democrat supporters didn't show up in the numbers expected.  And the diehard MAGA demographics voted at an unprecedented rate in 2016.

Those same pollsters got it wrong again, in 2020 (Freudian slip: 2024).  They have never gotten Trump's poll numbers right, so I am suspicious of them doing so now.  A +1% Harris lead in swing states could easy fall to Trump for that reason.

Between Donald "I didn't lose 2020" Trump and Joe "What year is it?" Biden, I disliked both choices.  Harris is an improvement.

On a slightly related note, the Federal Reserve just made a 0.50% rate cut, rather than a 0.25% cut.  As the economy slows, the Fed is lowering rates to encourage economic activity (and also to get back to a normal level).  The larger cut will help the economy, which could help Harris.  Expect Trump to complain about it.

I agree that it's possible Trump's support is understated again.  This is where my prediction might fall short.  I generally recall that the pollsters did a better job in 2020 than 2016, but I've honestly forgotten the details.  But these pollsters are always correcting their models based on more recent elections.  So I don't think this will be a major factor, but we need an election to find out. 

If I have to handicap it, I'm still going to say Trump voters are probably marginally less likely to show up at the polls than a statistical model predicts and Harris voters marginally more likely.

I also wonder how the down-ballot races are going to play out.  While I don't have any specific predictions, I think there will be some surprises based on turnout in various counties/states.  I think a few legislators in "safe seats" may be packing up their bags.  I suspect part of the surprise-factor on voter turnout will vary heavily by geography.  Maybe an abortion amendment will make more people show up in some states.  Maybe the people in Springfield OH vote in overwhelming numbers because of the national attention (on both sides).  Down-ballot issues will play a big role in how motivated voters are.

Like you, I recall the undercounting was reduced in 2020 but not eliminated.  I also don't know how much that varied across states, which matters when counting swing states.

I believe voters polled on the economy are warming up to Harris, who has reduced the gap with Trump on that issue.  The Fed cutting rates 0.5% takes some pressure off the economy, so that warming trend may continue.

I suspect the residents of Springfield are fed up with problems caused by immigrants.  Haitians are by all accounts hard working, and making an important contribution to the local economy.  But what residents see are bad driving and the spread of disease - both owing to Haiti's lack in those areas.  It strikes me as ironic that JD Vance is both complaining about the problems of these immigrants in Ohio, while being a senator from Ohio!  The governor of Ohio claims Federal assistance is needed - and that seems like something JD Vance should have been working on.  I suppose in the Vice Presidential Debate, that will be one of the topics.

[reeshau]

FiveThirtyEight actually archives the prior elections, for anyone who wants to delve into the history.

https://projects.fivethirtyeight.com/2016-election-forecast/

https://projects.fivethirtyeight.com/2020-election-forecast/

[Ron Scott]

Supposedly—the economy, immigration, and abortion are 3 big issues that Americans focus on now.

I’m not sure there’s anything left to say about abortion but Harris needs to keep the issue front and center and continually tie Trump to Roe and a national ban. Lotsa money in TV ads. Personal horror stories from young women, etc. Maybe they can get a few people off the couch?

Rs will continue to poke around for immigration horror stories and make them up if they can’t find any. Ds will complain. Lotsa R TV ads here. No side-switching potential. Couch potential.

The Fed half-point cut with the promise of another 50 bps by year end could move the markets. Another month of low inflation could help too. Not a lot of help by Nov for the American Dream tho.

Bottom line, I don’t see needle moving opportunities in these big 3. I suppose you could make the case that Trump and JD will continue to act like dicks and lose half a point there, I don’t know.

October surprise? There’s nothing on the horizon I can see but I imagine the Trump camp tries something, and maybe their international supporters will too.

[reeshau]

October surprise? There’s nothing on the horizon I can see but I imagine the Trump camp tries something, and maybe their international supporters will too.

Well, Melania's memoir comes out October 8.  The idea she would be dishing seems outlandish, but the rollout of it has been...weird.  less than 3 weeks to go, and no interviews, previews, etc.  It also has a stark black and white cover.  If it was fluff about modelling, fashion, and decorative items, you would think the cover would reflect some of those.

[LennStar]

The problem I have with the US economy—which is not the Fed’s fault—is with a lackluster growth in productivity. Productivity gains are magic, enabling more money to circulate in the economy without a concomitant increase in inflation…unless they’re low, which is where we’re at. I’d love the politicians to focus on this, but I’m not holding my breath.

Productivity increases happen if there is either labor shortage or high labor cost. Then automation becomes appealing.
So if you want higher productivity, lobby for a higher minimum wage, pro abortion and no immigrants.

Though of course other society things may also be a factor. Greeters in Japanese stores for example.

[Michael in ABQ]

In regard to polling errors, 2020 and 2016 both underestimated the percentage of votes for Trump by a significant margin. In 2020 this was consistent across almost every poll from those taken immediately before the election to those taken two weeks out. I've attached a histogram showing this - it's from Page 19 of this report by the American Association for Public Opinion Research (AAPOR).

https://aapor.org/wp-content/uploads/2022/11/AAPOR-Task-Force-on-2020-Pre-Election-Polling_Report-FNL.pdf

● The 2020 polls featured polling error of an unusual magnitude: It was the highest in 40 years for the national popular vote and the highest in at least 20 years for state-level estimates of the vote in presidential, senatorial, and gubernatorial contests. Among polls conducted in the final two weeks, the average error on the margin in either direction was 4.5 points for national popular vote polls and 5.1 points for state-level presidential polls.

● The polling error was much more likely to favor Biden over Trump. Among polls conducted in the last two weeks before the election, the average signed error on the vote margin was too favorable for Biden by 3.9 percentage points in the national polls and by 4.3 percentage points in statewide presidential polls.

● The polling error for the presidential election was stable throughout the campaign. The average error matched closely for polls conducted in the last two weeks, in the final week, and even in the final three days. The challenges polls faced in 2020 did not diminish as Election Day approached..

Here's another analysis with a graphic showing the difference between polling and actual results at the national level.
https://www.pewresearch.org/short-reads/2024/08/28/key-things-to-know-about-us-election-polling-in-2024/


However, since the national popular vote is a vanity metric that don't actually determine the winner, here is an analysis of swing-state polls in 2016 and 2020
https://centerforpolitics.org/crystalball/polling-error-in-2016-2020-look-out-for-wisconsin/

In 2016 the average swing-state polling overestimated Clinton by 2.5% - with only one state overestimating Trumps vote percentage.


In 2020 the average swing-state polling overestimated Biden by 2.4% - with only one state overestimating Trumps vote percentage (by 0.1%) and almost every single state overestimating Biden's voting percentage by at least 1.5%.

[dandarc]

Huh? "In 2016 the average swing-state polling overestimated Clinton by 2.5% - with only one state overestimating Trumps vote percentage."

Tracks - average the last column is 2.51% towards dems.

"In 2020 the average swing-state polling overestimated Biden by 4.5% - with only one state overestimating Trumps vote percentage (by 0.1%) and almost every single state overestimating Biden's voting percentage by at least 1.5%."

doesn't track - average of last column is 2.42% towards dems, not 4.5%. So by this dubious analysis of 7 states aggregate data, the pollsters got slightly better - not worse - from 2016 to 2020.

[Michael in ABQ]

Huh? "In 2016 the average swing-state polling overestimated Clinton by 2.5% - with only one state overestimating Trumps vote percentage."

Tracks - average the last column is 2.51% towards dems.

"In 2020 the average swing-state polling overestimated Biden by 4.5% - with only one state overestimating Trumps vote percentage (by 0.1%) and almost every single state overestimating Biden's voting percentage by at least 1.5%."

doesn't track - average of last column is 2.42% towards dems, not 4.5%. So by this dubious analysis of 7 states aggregate data, the pollsters got slightly better - not worse - from 2016 to 2020.

Corrected. Not sure how I messed up the average when I was adding them up. It should have been a 2.4% difference not a 4.5% difference in 2020.

[Telecaster]

In regard to polling errors, 2020 and 2016 both underestimated the percentage of votes for Trump by a significant margin.

Yes, but with a caveat.  The margin of error for a typical poll is usually about 3-3.5%.    So individually the polls were generally within the expected margin of error.  Using polling averages should tend to cancel out the errors and converge on the "true" result.    The surprising part was the polls were all wrong in the same direction.   Which means there were (and probably still are) systemic problems with the polling process.   

[nereo]

Huh? "In 2016 the average swing-state polling overestimated Clinton by 2.5% - with only one state overestimating Trumps vote percentage."

Tracks - average the last column is 2.51% towards dems.

"In 2020 the average swing-state polling overestimated Biden by 4.5% - with only one state overestimating Trumps vote percentage (by 0.1%) and almost every single state overestimating Biden's voting percentage by at least 1.5%."

doesn't track - average of last column is 2.42% towards dems, not 4.5%. So by this dubious analysis of 7 states aggregate data, the pollsters got slightly better - not worse - from 2016 to 2020.

Corrected. Not sure how I messed up the average when I was adding them up. It should have been a 2.4% difference not a 4.5% difference in 2020.

To be clear, are you saying that because the polls underestimated vote percentage for Trump in 2016 & 2020 that it is likely the same thing will happen in 2024?

[Michael in ABQ]

In regard to polling errors, 2020 and 2016 both underestimated the percentage of votes for Trump by a significant margin.

Yes, but with a caveat.  The margin of error for a typical poll is usually about 3-3.5%.    So individually the polls were generally within the expected margin of error.  Using polling averages should tend to cancel out the errors and converge on the "true" result.    The surprising part was the polls were all wrong in the same direction.   Which means there were (and probably still are) systemic problems with the polling process.

I came across a really good article a week or two ago that went more in depth on the comparison between the Nate Silver 538 weighted average polling method vs. the Real Clear Politics simple average polling method. Unfortunately, I can't seem to find it again. The analysis and conclusion were similar. 2016 and 2020 the polls consistently underestimated Trump at both the state and national level by a couple of percent. Some states the simple average was closer than the weighted average or vice versa, but in both cases the polls as a whole consistently missed in the same direction.

[MustacheAndaHalf]

Michael in ABQ - you can append "#page=19" to the url of a PDF to automatically go to page 19.  But I think national polls aren't useful for predicting the electoral college - only swing states matter.  Which is why I find page 32 of your link interesting: specific polling error by state.

Wisconsin (WI) +8% error
Pennsylvania (PA) +4% error
Michigan (MI) +4% error
https://aapor.org/wp-content/uploads/2022/11/AAPOR-Task-Force-on-2020-Pre-Election-Polling_Report-FNL.pdf#page=32


I recall Quinnipiac polls not having great reliability, but it has the advantage of being from Thursday, Sept 18.  It doesn't really matter if I use the same polling error, half the error, or even 1/4th.  Harris is ahead far enough in two swing states, and not enough in another:

PENNSYLVANIA: Harris 51%, Trump 45%, Stein 1%, Oliver 1%
MICHIGAN: Harris 50%, Trump 45%, Stein 2%
WISCONSIN: Harris 48%, Trump 47%, Stein 1%
https://poll.qu.edu/poll-release?releaseid=3905

If you push Wisconsin's votes +2% +4% or +8% in favor of Trump, he wins the state.  Prior polling errors suggest it is not a tie.  But similar adjustments (+1% +2% +4%) aren't enough to make up for Harris' 5% and 6% leads in the other states.  There are other swing states not covered by this recent Quinnipiac poll.

[NorCal]

Huh? "In 2016 the average swing-state polling overestimated Clinton by 2.5% - with only one state overestimating Trumps vote percentage."

Tracks - average the last column is 2.51% towards dems.

"In 2020 the average swing-state polling overestimated Biden by 4.5% - with only one state overestimating Trumps vote percentage (by 0.1%) and almost every single state overestimating Biden's voting percentage by at least 1.5%."

doesn't track - average of last column is 2.42% towards dems, not 4.5%. So by this dubious analysis of 7 states aggregate data, the pollsters got slightly better - not worse - from 2016 to 2020.

Corrected. Not sure how I messed up the average when I was adding them up. It should have been a 2.4% difference not a 4.5% difference in 2020.

To be clear, are you saying that because the polls underestimated vote percentage for Trump in 2016 & 2020 that it is likely the same thing will happen in 2024?

I know a lot of people are building a mental model of that by taking current polls and adding the polling error for last election.

I’m no statistician, but even I know that’s not how it works.

Demographics have changed, support among different groups have changed. Some percentage of the population that voted in 2016 (10%ish?) is dead.  8 years of high school graduates that couldn’t vote in 2016 can now vote.  Polling firms are constantly updating their models to adjust to any known bias.

Trying to pretend there’s a predictable and simplistic systematic bias is just intellectual laziness.

It doesn’t mean the models won’t swing within the margin of error. Assuming they’re using a 95% confidence interval, there’s a 95% chance the polls land within their margin of error on either side of the equation. That’s a pretty wide range.

[Ron Scott]

Trying to figure out how the candidates are planning to deal with energy production and climate management is tricky.

Trump is most likely burn baby burn, but I’m sure Elon’s got him soft on electric cars.

Harris has promised not to interfere with fracking so I guess she’ll keep that promise.

Meanwhile—while renewables are getting better and cheaper, fossil is still strong and probably growing. And AI/big tech are planning mega-energy usage wherever they can get it.

FWIW—Bill Gates just bought all the production for 20 years on a 3 Mile Island restart. Fun!

[ChpBstrd]

Demographics have changed, support among different groups have changed. Some percentage of the population that voted in 2016 (10%ish?) is dead.  8 years of high school graduates that couldn’t vote in 2016 can now vote.  Polling firms are constantly updating their models to adjust to any known bias.

Trying to pretend there’s a predictable and simplistic systematic bias is just intellectual laziness.

Speaking of people dying, COVID killed 1.2 million Americans, the vast majority of whom had been citizens eligible to vote in 2016. 70% of these deaths were in people over 70, a historically Republican demographic. Even if we naively assume no difference in risky behavior or odds of fatality between Republicans and Democrats during the pandemic, the characteristic of the virus to kill elderly people much more often than young people can be expected to reduce Republican votes. Last week, COVID-19 accounted for 2.3% of deaths in the United States.

That said, life expectancy in the U.S. is back up, with more middle-aged people each day entering their elder years when conservative ideas seem to be more appealing. America got younger in 2020-2021, but then got older in more recent years.

An insight I gathered from the debate is that elections are now more about epistemology than ideology. Trump does not even resemble Ronald Reagan or George H.W. Bush ideologically, but rather appeals strongly to people who get their news and information from YouTube or X. These sources have become like Fox News in many ways. YouTube's algorithm has been experimentally proven to direct users toward right-wing content. Twitter, meanwhile, was purchased by a billionaire for the expressed purpose of making it a right-wing platform.

We can see that these platforms have changed the ideology of Americans over the past 4 years. Just look at vaccine confidence and trust in medical professionals. A 2024 study found that Americans are actually becoming less confident in COVID-19 vaccines and information from expert sources like the FDA and CDC.

Now, who could be doing that? This is the work of Elon Musk and the Board of Alphabet. For everyone not in the right-wing media bubbles, receiving their information from mainstream media or government websites, the war on COVID was won through vaccines, masks, and eventual herd immunity at a great cost in lives. For those getting their info from social media though, COVID was always a scam by the elites to make money and take away people's freedom, and the "Wuhan flu" was possibly never a real disease.

What you believe is literally a matter of where you get your information, and today's red/blue divide simply reflects what information sources people are willing to trust (i.e. it is epistemology for information consumers). 

Another example: 56% of Americans thought the economy was in recession in the 2nd quarter of 2024. The Bureau of Economic Analysis, on the other hand, reported blockbuster GDP growth of 3.0% for that same timeframe. Obviously, some people were getting inaccurate information about the external world. Who might be doing that? If you believe the influencers, the government is printing fake numbers to deceive us. If you believe the BEA, the people following influencers on X and YouTube are in an information bubble.

So Trump may have looked stupid to Blue America when he repeated internet conspiracy theories and called the economy horrible, but to Red America he is just repeating their news - and being brave enough to do so under hostile questioning from the untrustworthy "lamestream" media elites who want to cover things up for the Biden/Harris regime.

The two types live in different worlds, with different truths and different experiences of reality. Whatever your position on this spectrum, the smart thing to do right now is look for the disconfirming evidence with your own eyes rather than assuming the other side is foolish. E.g. If you feel like the economy is booming, ask yourself how all those people begging on street corners got there, or why young people feel they can't start families. If you feel like the economy is actually in recession, hop onto a jobs website like Indeed or LinkedIn to have a look around your local area. Do this for everything... immigration, whether particular groups of people are oppressed or not, etc. The best foreign policy research is done by travel, not googling.

[LennStar]

FWIW—Bill Gates just bought all the production for 20 years on a 3 Mile Island restart. Fun!

Not BG but MS, to power their AI.
Because 3 of those are not enough for good results.

[MustacheAndaHalf]

...
To be clear, are you saying that because the polls underestimated vote percentage for Trump in 2016 & 2020 that it is likely the same thing will happen in 2024?

Demographics have changed, support among different groups have changed. Some percentage of the population that voted in 2016 (10%ish?) is dead.  8 years of high school graduates that couldn’t vote in 2016 can now vote.  Polling firms are constantly updating their models to adjust to any known bias.

Trying to pretend there’s a predictable and simplistic systematic bias is just intellectual laziness.

Although you only mention 2016, pollsters tried to correct the bias in 2020 and failed again.  You mention demographic changes, but that isn't the problem.  Trump voters are less willing to talk to pollsters and/or admit they're voting for Trump.  The political divide has cut families in two, leaving Trump voters isolated from Biden voters.  There are strong pressures for Trump voters to hide their politics, and I don't think a random phone call overcomes all of that pressure, which creates different response rates for Trump voters.

The bias was factored in for 2020, and did come in closer than 2016.  I assume 2024 will have an even narrower bias than the prior two elections.  But dismissing that bias is a very risky thing to do in an election this close.

[Telecaster]

Demographics have changed, support among different groups have changed. Some percentage of the population that voted in 2016 (10%ish?) is dead.  8 years of high school graduates that couldn’t vote in 2016 can now vote.  Polling firms are constantly updating their models to adjust to any known bias.

Trying to pretend there’s a predictable and simplistic systematic bias is just intellectual laziness.

It doesn’t mean the models won’t swing within the margin of error. Assuming they’re using a 95% confidence interval, there’s a 95% chance the polls land within their margin of error on either side of the equation. That’s a pretty wide range.

There does seem to be systemic sampling error however.   The confidence interval assumes the samples are representative of the population.   If that were the case, then the averages should center around the "true" value.   However, the 2020 polls had high margins of error and errors were in one direction.    There are many theories why this was the case, but no one seems to know for sure or how exactly to fix it.   

Most polling averages show Harris with a very slight lead.  This is terrifying to me because unless the sampling errors have been fixed Trump will be the likely winner. 

[partgypsy]

I do know. I work at the VA and past year needed to call vets or caregivers for research. And my very unscientific observation it seems that people are more suspicious about the government and research than even a few years ago, in the sense of a lot more hang ups before I can finish a sentence. Or maybe people are less polite. It's hard to understand how accurate a phone poll is when so many people do not answer their phone.

[Tigerpine]

I do know. I work at the VA and past year needed to call vets or caregivers for research. And my very unscientific observation it seems that people are more suspicious about the government and research than even a few years ago, in the sense of a lot more hang ups before I can finish a sentence. Or maybe people are less polite. It's hard to understand how accurate a phone poll is when so many people do not answer their phone.

I don't distrust the government so much as I distrust being called by someone I don't know.  I won't even pick up the phone 9 times out of 10.  And nowadays scammers can mimic other phone numbers on caller ID, so that's not even a dependable filter anymore.  I would be one of those to hang up on you if I did happen to pick up.  It's no offense to you.  There are just too many dishonest players out there.

[ChpBstrd]

I do know. I work at the VA and past year needed to call vets or caregivers for research. And my very unscientific observation it seems that people are more suspicious about the government and research than even a few years ago, in the sense of a lot more hang ups before I can finish a sentence. Or maybe people are less polite. It's hard to understand how accurate a phone poll is when so many people do not answer their phone.

I don't distrust the government so much as I distrust being called by someone I don't know.  I won't even pick up the phone 9 times out of 10.  And nowadays scammers can mimic other phone numbers on caller ID, so that's not even a dependable filter anymore.  I would be one of those to hang up on you if I did happen to pick up.  It's no offense to you.  There are just too many dishonest players out there.

So by extension, maybe the proliferation of auto-dialers and phone scams in the last 8-10 years has reduced trust/patience in unfamiliar callers, like pollsters. And perhaps this effect is more pronounced among people who would vote for Trump because of some reason such as:

a) they get more such calls (possibly due to being a scam victim before?)
b) they are more likely to give out their phone number to companies that sell their phone number,
c) they are not as good at screening these calls, or they get more annoyed by these calls than non-Trump voters.
d) your guess?

[bacchi]

Demographics have changed, support among different groups have changed. Some percentage of the population that voted in 2016 (10%ish?) is dead.  8 years of high school graduates that couldn’t vote in 2016 can now vote.  Polling firms are constantly updating their models to adjust to any known bias.

Trying to pretend there’s a predictable and simplistic systematic bias is just intellectual laziness.

It doesn’t mean the models won’t swing within the margin of error. Assuming they’re using a 95% confidence interval, there’s a 95% chance the polls land within their margin of error on either side of the equation. That’s a pretty wide range.

There does seem to be systemic sampling error however.   The confidence interval assumes the samples are representative of the population.   If that were the case, then the averages should center around the "true" value.   However, the 2020 polls had high margins of error and errors were in one direction.    There are many theories why this was the case, but no one seems to know for sure or how exactly to fix it.   

Most polling averages show Harris with a very slight lead.  This is terrifying to me because unless the sampling errors have been fixed Trump will be the likely winner.

The 2022 polls were very accurate but then the House and Senate polls for 2020 were more accurate than the Presidential race.

However, the Presidential polls weren't that far off in 2016/2020 compared to the past. Look at the 1980 national polls, which were off by 8 points (R), or 2000 which was off by 5 (D).(1)


(1) https://fivethirtyeight.com/features/the-polls-werent-great-but-thats-pretty-normal/

[bacchi]

I do know. I work at the VA and past year needed to call vets or caregivers for research. And my very unscientific observation it seems that people are more suspicious about the government and research than even a few years ago, in the sense of a lot more hang ups before I can finish a sentence. Or maybe people are less polite. It's hard to understand how accurate a phone poll is when so many people do not answer their phone.

I don't distrust the government so much as I distrust being called by someone I don't know.  I won't even pick up the phone 9 times out of 10.  And nowadays scammers can mimic other phone numbers on caller ID, so that's not even a dependable filter anymore.  I would be one of those to hang up on you if I did happen to pick up.  It's no offense to you.  There are just too many dishonest players out there.

So by extension, maybe the proliferation of auto-dialers and phone scams in the last 8-10 years has reduced trust/patience in unfamiliar callers, like pollsters. And perhaps this effect is more pronounced among people who would vote for Trump because of some reason such as:

a) they get more such calls (possibly due to being a scam victim before?)
b) they are more likely to give out their phone number to companies that sell their phone number,
c) they are not as good at screening these calls, or they get more annoyed by these calls than non-Trump voters.
d) your guess?

d) Older and taught to answer every call.
e) A slight persecution complex, like we've seen on social media, where they fear expressing their views will get them criticized or ostracized.