Reverse engineering Bowling Alone

After reading Robert Putnam's Bowling Alone, I'm trying to suss out a few things. First, he attributes changes in demographic values as the main cause of the decline in social capital: the Silent Generation cared more about things like volunteering, voting, and philanthropy; Boomers less so, Gen X even less, Millenials hardly at all.

But why? Also, is there anything else he missed?

In The Big Sort, Bill Bishop has some interesting observations.

"...large partisan majorities had the effect of dampening turnout among citizens in the political minority. Rather than buck the majority and risk social sanction, citizens in the minority simply stayed away from the polls.

In communities with large political majorities, people tend to give up battling over ideas.... A vote becomes more an affirmation of the group than an expression of a civic opinion."

Bishop notices the trend of people moving into like-minded communities, or what he calls sorting. If what he writes is true, groupthink and demographic shifts by community are major causes of the decline in civic engagement that Putnam has studied. So is sorting a cause of the decline in voting and other forms of social capital?

Kahneman and Tversky showed that loss aversion is the primary motivating factor for human behavior. Maybe that is what drives people to vote, write to their political representative, attend town hall meetings, or any of the other things Putnam used to measure civic engagement: they feared losing their way of life.

Maybe they stay home now because, due to sorting, their ingroup dominates their community and is getting their way. They don't feel the need to make their voice heard. Or, as in the quote above, they are a political minority not wanting to risk social isolation.

Of course, Putnam also said that more people are identifying as moderates and the extremists are the ones participating in their communities. If that's true, then extremism and social capital would be correlated. Are they?

I'm the map I'm the map I'm the map

Putnam insists that social capital improves tolerance. (Maybe he was just looking at tolerance for minorities and LGBT+?) However, if you look at the social capital index map of the U.S. and compare it to The Atlantic's map of political prejudice, it doesn't say what you would expect it to.

Massachusetts and Connecticut, which are in the top 60-80% for social capital, are among the least politically tolerant places in the country. New York, which is in the bottom 20% for social capital, is probably the most politically tolerant state.

Florida is disappointing in both regards, which is the most Florida thing ever. And North Dakota seems to score well in both measures.

Amanda Ripley's story declares Jefferson County New York the most politically tolerant county in the U.S. (although the interactive map places it in the 3rd percentile, where lower is better). Meanwhile, the social capital index places Jefferson County in the 34th percentile (where higher is better), scoring particularly poorly in Institutional Health (voting, confidence/trust) and not much better in Community Health (volunteering, church attendance, public meetings, etc.).

(The Atlantic's methodology could be the answer for the disparity. They excluded independents from their data. So a county with a high number of independent voters would only look at Rs and Ds, which might not be an accurate reflection.

Also, they relied on a survey. The social capital index looks at a lot of data, including measures of behavior, which I trust more. I'm more likely to believe a church's records of attendance than asking people to tell me how often they go to church.)

Still, I can't get over how different the results are. Maybe social capital isn't a strong predictor of political tolerance. Maybe it's political equilibrium.

(edit: super interesting study here finds that "when answering questions about the other party, individuals think about elites more than voters, and express more animus when the questions focus on elites. This suggests that increased affective polarization reflects, to some extent, growing animus towards politicians more than ordinary voters..."

"Fully 80% of our sample is “somewhat comfortable” or “extremely comfortable” with being friends or neighbors with those from the other party. While comfort with inter-party marriage is lower, even here, there is less anger than one would expect, with only 5% of the sample being “extremely upset” by this. This underlines that while people might feel negatively about the other party in the abstract, they are still comfortable interacting with them in ordinary situations."

So the politically tolerant counties might just overwhelmingly be politically disengaged people who don't follow national news. And politically intolerant counties answered the survey with U.S. Senators in mind, not their neighbors.)

Competition Breeds Collegiality? 

Let's take a look at two maps. This first one is from The Big Sort showing how counties voted in the 2004 presidential election. White counties are "competitive," meaning the winner captured fewer than 20% of the vote.

Here is FiveThirtyEight's map from the 2016 election. Likewise, white counties are where the winner won by fewer than 20%.

So far, Maine, New Hampshire, Arizona, Colorado, Florida, North Carolina, New York, Virginia, Wisconsin, Michigan, Minnesota, Connecticut, and Washington all seem to have counties with less "sorting" and a more even distribution of Rs and Ds.

Now here is the political tolerance map:

Florida and Nevada seem to be opposites, but New York and North Carolina show a correlation with the 2016 election map. You would expect the white counties to match up better; places with an even number of R and D voters should be more tolerant of one another. But that doesn't seem to be the case.

And here is the Social Capital Index map, where cream and light green counties are the highest (best):

New England and some of the Great Lake states, and maybe Colorado seem to match well with the 538 map of competitive counties, but not much else. Maybe the moderates and politically disengaged are sorting into similar counties and not voting or participating civically, since they are not activists.

So there is some correlation between the three, but not much. Of course, this is just me eyeballing the map. Ideally I'd run a linear regression model. But, ya know, math.

Hierarchy of Needs

In the 1970s Political Scientist Ronald F. Inglehart seemed to predict the decline in social capital, well before Bowling Alone was published in 2000. Putnam was convinced that change in generational values was the largest factor for the decline in social capital, but Inglehart goes into the reasons why.

Sebastian Junger once said, "No food, one problem. Lots of food, many problems."

The Great Depression and WWII made hunger and security a problem. So the Silent Generation generally got along once we got out of that mess. As we became more prosperous in the 60s, people began caring about self actualization (ie "many problems").

So why didn't 911 and the Great Recession mirror the tragedies of the 30s and 40s and lead to bipartisanship? It's hard to tell the years below, but it seems like there was some stability after 9/11, but unfavorable views among the parties increased as we went into the recession.

The recession did seem to make things worse, according to this article.

" But a new study in the October issue of the American Sociological Review found that public support for government efforts to address social problems actually declined in the wake of the 2008 economic crisis.

"We found it surprising that as the Great Recession emerged, the American public moved quickly toward lower levels of support for government policy solutions to social problems," said Clem Brooks...

"The key is partisanship," Brooks said. "We found that it wasn’t the public as a whole that moved away from support for government policy solutions. Self-identified Democrats actually became slightly more supportive of government solutions between 2008 and 2010. But self-identified Independents and especially Republicans experienced a large shift in the opposite direction." 

This would agree with the unfavorable ratings among Rs being worse, according to the Pew chart above. It should surprise no one that Democrats supported the government when they controlled all three branches and Republicans opposed it.

Did I made a huge mistake?

As a localist, the question I'm struggling with is: Is that what I want; people voluntarily moving into ideological bubbles even if it makes them more extreme?

I believe that people should be able to live in a community of like-minded citizens. And I believe it should be easy for people to leave a community where they don't fit in to one where they do. (And I believe that top down, one size fits all solutions make this type of living impossible.)

I worry that this direction will lead to extremism at the local level. But there may be a split where people seek communities that prioritize civil behavior over extremism.

Finally,

Both Bishop and Putnam seem to suggest that extremists are more active politically, and moderates less so. Putnam says the share of self-identified moderates is increasing. I just don't know where this is showing up in the data.

I guess it could be that increasing landslide victories isn't just indicative of sorting, but of moderates staying home and not voting. And those moderates might be politically tolerant people, so the extremists who do vote overwhelmingly for a candidate aren't representative of the entire county.

I still want to know what causes politically tolerant counties? Have they been unaffected by sorting? Are they filled with moderates? Do they have lots of interaction with their outgroup?

This data makes me less hopeful that social capital can decrease polarization, especially if sorting, loss aversion, and a common enemy are what drive social capital.

I think the story is that social capital was highest in the 1950s and political polarization and tolerance was doing well too. Social capital didn't just decline in the 1960s, it became concentrated among extremists who started living near one another, creating a feedback loop of moderate disengagement.

I believe that social capital works best when moderates have a seat at the table. How to encourage that is the question I'm trying to answer.

The New Civic Engagement

I.
This article in The Atlantic about the Wayfair Walkout reminded me of Robert Putnam's work in Bowling Alone. Putnam noted the decline in social capital by looking at civic engagement, including people who write to their political representative or attend a rally.

All of these things have been declining since the 1950s, but maybe he's just looking in the wrong area. Maybe people's distrust of government has only directed their frustration from politics to private business.

From the article:

"This is a lesson that the large tech platforms have learned, in no small part because their employees have begun to hold them accountable. In the last year, workers at Microsoft, Salesforce, Google, and Amazon have all pressured their employers to revisit government contracts they deemed unethical. Now, it seems, less overtly political companies are also learning that every decision has political valence."

II.
It's weird to think that we're living in a society in which consumers are expecting businesses to reflect their values. This reminds me of Jeremy Heimans and Henry Timms when they write about New Power models:

"Patagonia, for example, has a traditional old power business model, yet it stands out for its embrace of new power values like transparency. Some of these “cheerleader” organizations, such as The Guardian newspaper, are working to evolve their positions so that they not only espouse new power values but incorporate new power models effectively."

A conservative example would be the people who burned their Nike shoes after the company made Colin Kapernick the face of their new campaign.

III.
I work in higher education. My role is in fundraising. Most of our donors are Baby Boomers. In the coming years, a significant portion of our donors and alumni will be Millenials.

Eventually, our institution is going to be forced to take a side on some controversial scissor statement, something that ferociously divides people along sharp sectarian lines. Maybe we'll be expected to give hiring preferences to veterans. Maybe we'll be expected to lower admission standards for minorities. Maybe we'll be expected to have a carbon neutral footprint.

And those Millenials are going to want to know we are on the "right" side.

I don't think we're prepared for that. I know trying to placate both sides is futile. The question becomes: does choosing a side result in loyalty (donations) that is commensurate with disaffecting the faction we don't side with?

The Newton Street Plaza Problem

The three owners of the Newton Street Plaza are proposing a housing complex.

According to research into the Costs of Community Services:

The average estimate ranges (for residential land development) from about 1.15 to 1.50, which means that for every dollar collected in taxes and non-tax revenue, between $1.15 and $1.50 gets returned in the form of local government and school district services.... 

According to the COCS studies, the largest single expenditure category for communities is the public school system, accounting for 61.4 percent of spending. Since open space and commercial development in themselves do not place any burden on the schools, it should not be surprising that their ratios are lower than those for the residential category.... 

If many homes in a community are in an extremely high price range and occupied by “empty nesters,” for example, the COCS ratio should be expected to be relatively low. On the other hand, low- or middle-income property occupied by families with numerous children would produce a higher ratio.

In short: this will increase the number of people who will require local services (schools, police, fire, roads, but mostly schools) without adding people paying local property taxes to offset those services. The owners will pay some property tax, and the residents' spending will add to the local economy, but not enough to offset anything.

What we have is a misalignment of incentives. The owners get to make money but not have to deal with the town's net deficit and overcrowded schools. Other than Yee, no one else lives in town. And unless he has children in the SH school system, he won't be affected by the increased load there.

I don't want to be anti-growth; I just want to be smart about it. I'd like to reduce the incentives for these types of residential developments and increase the incentives for tax-surplus developments (businesses and open spaces).

I. Discourage new rental housing (without skin in the game)

Any new multi family property development must be owned by someone with a kid in the SH school system. The idea is to create a disincentive for owners to invest in an arrangement in which they are not bearing the externalities. The largest local expenditure is schools; so they should have skin in the game.

This won't stop these developments or add more revenue, it will just force developers to have skin in the game. More consideration will be put into the whole process if they share the burden with the town.

Commercial/industrial space, or open land/agricultural space, can be left alone since the research shows that they are a net surplus on local government.

II. Offset Residential Development with Net Surplus Development

Encourage commercial growth
Others have argued that these structures must have a business on the first floor. That makes sense. According to the same research:

For commercial/industrial, the ratio usually ranges from 0.35 to 0.65, indicating that for every dollar collected, the local government provides only about 35 to 65 cents worth of services. For agriculture and open space, the ratios are only slightly smaller, usually ranging from 0.30 to 0.50.

How about this: a new residential development cannot be approved until a new commercial or open space development is approved. A really anxious owner can contribute to this project if they want to get their housing started.

Attract/Retain empty nesters
Empty nesters do not add to the school burden. We should do everything we can to attract and retain this population. They want low crime, low congestion, walkable communities, and probably a good senior living center. According to this: dog parks, lectures (hello MHC), community gardens, and low maintenance/high end homes are what retirees look for. Keep these options in mind.

More Parks?
Although it doesn't make intuitive sense, the research shows that developing a public park is a net benefit for a town. It doesn't add any revenue and has some expense for maintenance and probably police patrolling for loiterers.

But apparently it adds enough to property values to bring in more revenue. I couldn't be happier. I'm a huge proponent of more parks, a great source of social infrastructure.

How about this: the Newton Street Plaza owners pick up the tab on the proposed dog park. Or, they make annual payments to subsidize the municipal golf course.

Final thoughts
When you restrict housing, you inevitably price people out of the market. This is what is happening on a larger scale in our largest growing cities. Lower end housing isn't just poor people, it's often young people, struggling with student debt, supporting their families, and dealing with entry level pay. They should have their shot at the American dream too.

The population keeps growing; we need to build housing to keep up with it. So I propose a 1:1 new development ratio, a skin in the game requirement for residential developers, and incentives to keep/attract empty nesters.

Civility vs. Activism

I.
I keep thinking about the two campaign strategies: try to turn non voters into voters or try to turn moderate voters into voting for you. One assumes that there are a lot of people who think like you and you just have to convince them to vote. The other assumes that you can convince people who aren't that different from you to switch sides and vote for you.

One thinks that people don't change and any effort to do so is wasted energy. The other thinks that people are complex and some of them can change if you engage with them.

I think whichever one you favor says a lot about you.

I like Ezra Klein. He is thoughtful and kind; he invites people he disagrees with onto his podcast. Which is why it shocks me when I hear him say things about how he prizes blunt activism over working with the other side.

I can recite dozens of reasons why my path of civility/moderation/compromise is better, but I don't think any of that matters. Because I don't think I get to choose.

I think my coercion aversion (I really need a better name for that) decides for me. I think I prefer civility and compromise because, at my core, I really am uncomfortable telling people what to do or think.

A libertarian economist like Russ Roberts can write a long essay about how minimum wage laws actually hurt low skilled workers more than they help them. But I don't know that he doesn't just feel uncomfortable telling businesses how much they have to pay people, and then worked to find a good reason to justify that feeling.

II.
I think higher education was dominated by the civility types. Now, the activists are having their say. Activism gets things done, while moderates hem and haw about the right approach.

I know the academic pursuit of truth calls for careful consideration about what is good and true, but it didn't seem to make much room for the academic activists.

Maybe this is like Jonathan Haidt's idea of colleges deciding to be Truth U or Social Justice U. Activist students should have the opportunity to go to a school where they will be taught how to mobilize for social change: conducting multi cultural training, organizing protests, boycotting non fair-trade companies, and so on.

Truth U will teach students how to consider numerous viewpoints, how easily we can be fooled by data, civil debate, gentle persuasion, and so on.

Maybe viewpoint diversity is just one viewpoint and it isn't for everyone. It's not enough to split the world into liberals and conservatives. There are also liberals and conservatives who don't want to engage with their outgroup and liberals and conservatives who do. Maybe they both need their own space.

That's easy enough for higher education, but what about politics? We only have one governing body and the battle between civility vs. activism seems as important as the battle between Democrats and Republicans.

III.
Maybe without realizing it, groups like Better Angels, Living Room Conversations, and More in Common—groups that seek to unite liberals and conservatives—are actually crystallizing a political faction that has not had a home in civic participation. Instead of bridging the divide between reds and blues, they are just building a home for the existing reds and blues who don't hate each other.

From Robert Putnam's Bowling Alone:

"Ironically, more and more Americans describe their political views as middle of the road or moderate, but the more polarized extremes on the ideological spectrum account for a bigger and bigger share of those who attend meetings, write letters, serve on communities, and so on. The more extreme views have gradually become more dominant in grassroots American civic life as more moderate voices have fallen silent."

The Hidden Tribes report notes that "Progressive Activists" account for 8% of Americans, "Devoted Conservatives" making up 6%. Meanwhile "Moderates" and "Politically Disengaged" combined to form 41% of all Americans.

In general terms, I think what distinguishes these moderates is that they place greater emphasis on civility instead of pushing through their agenda.  And for good reason.

In a Democratic, pluralistic society, you need civility to accomplish your agenda. Plus, moderates do not want to live in a society in which authoritarianism, coercion, or, God forbid, violence is the means through which a faction achieves its goals.

And the activist faction might be less stable. Scott Alexander noted:

" Long before a group can take over society, it reaches a size where it needs to develop internal structure and rules about interaction between group members. If you collect a bunch of people and tell them to abandon all the social norms like honesty, politeness, respect, charity, and reason in favor of a cause – then the most likely result is that when your cause tries to develop some internal structure, it will be overrun by a swarm of people who have abandoned honesty, politeness, respect, charity, and reason...

"someone who will be a jerk for you will be a jerk to you ... But more importantly if you elevate jerkishness into a principle, if you try to undermine the rules that keep niceness, community, and civilization going, the defenses against social cancer – then your movement will fracture..."

But really, I'm just a guy who doesn't like aggressive confrontation on an emotional level and I'm looking for a community of like-minded believers who can confirm my priors. That's how I ended up as a member of Better Angels.

Predictions for small populations are useless

An article in Psychology today about gender representation in STEM had the following quote:

"Let's start with a simple fact: Most women do not have the right aptitude to be professors at top STEM departments. This is unfortunate, perhaps, but it’s true. It’s also true, though, that most men don’t have the right aptitude! Only a small minority of people do. The phenomenon we’re trying to explain is not why half the population (men) can do it whereas half the population (women) can’t. Most of the population can’t, and of the tiny fraction who can, some are men and some are women. The only question is: Why is the tiny fraction of men working in STEM fields today somewhat larger than the tiny fraction of women?"

I'm less concerned with where the author goes from there than I am with one specific part, the part where he points out that only a small minority of people work in STEM.

I wonder how much time we waste with forecasting and predictive models for something that has such a small sample size, that it's probably impossible to distinguish signal from noise.

Let's say the military wanted to identify US citizens at highest risk for joining ISIS. They collect all this demographic information of who has joined ISIS. They find that muslim men who immigrate from Aleppo, Syria, aged 17-28, with a college degree, who follow specific Twitter accounts are at highest risk to join.

So what happens when their spy program identifies a person who checks all those boxes? How likely are they to join ISIS? Still unlikely since MOST PEOPLE DON'T JOIN ISIS! If the base rate of people who join is 1%, all that statistical data does is move it to 3%.

When college isn't hard enough

I.
Something seems backwards about the signalling effect of colleges to employers. Ideally, colleges want two things: a low admission rate and a high graduation rate, especially the former.

It is very difficult for employers to project how a potential employee will perform, so they use college degrees as a signalling method. If it's really hard to get into Harvard, they must be exceptional.

However, this tells us exactly nothing about the quality of a Harvard education. 86% of freshman will graduate in four years. What about Oglala Lakota College, which has a 2% graduation rate? How exceptional do those handful of students have to be? It's hard to tell since it has a 100% acceptance rate.

Either way, isn't a low graduation rate a better signal of how difficult the experience is?

II.
For those who can't get into a low acceptance rate college, a better way to signal to employers might be to do something else that most people cannot do. Design your own app or website. Successfully run a small business. Build a combustion engine. Hell, ace a few MOOCs at Ivy League schools.

Imagine you're an employer with a stack of job applications for an entry-level position. They all have young graduates who went to the same schools, were president of the same clubs, and volunteered their spring breaks in the same third world countries. They also lack the hard skills to do the job, but that's fine since it's entry-level and you'll teach them anyway.

Then one resume comes from a student without any of those credentials. They graduated high school, then trained for a year before successfully climbing Mount Everest. They pledge to apply the same grit and determination to learning to crunch numbers for your firm. You don't know much about the other graduates' college experiences, but you know how few people have climbed Mount Everest. Is this a better candidate?

Maybe all the traditional signals—SATs, GPA, volunteering, extracurriculars, class ranking, internships—are becoming anti inductive; everyone does them now so you no longer stand out.

What if we told kids to do something demonstrably difficult, that most of your peers cannot/will not do, and it makes you stand out? Isn't that a better signal?

Social Signaling and Why Drugs are Cool

There is a scene in Swingers in which Jon Favreau is explaining to Ron Livingston the nature of Los Angeles clubs. From memory:

"For some reason all the cool bars in LA have no signs. It's like a speakeasy-sort-of thing. You tell a girl you went somewhere, it's like you're bragging you had to find it."

I think about this when I think about young people using drugs and alcohol. Sure, some people just enjoy being high. But I think something else is going on.

Evolutionary psychology posits that the thing women find most attractive in men is access to resources (this is a better description than money, since money didn't always exist.) Since most young people are unable to work enough to really signal how much money they have, they must find other ways to distinguish themselves.

In our society, access to resources comes in the way of being able to navigate one's social circle. In order to find drugs and alcohol, you have to know whom to ask. You have to know where the keg party is on Saturday. You have to know someone old enough to buy you beer.

The harder the resource is to find, the stronger your signal will be.

Telling a girl you bought the best bud in town, it's like you're bragging you had to find it.

My theory is that drugs are popular among young people because they signal one's access to resources. The best way to keep kids away from drugs is to replace their social signaling system with a new, less harmful one.