Just how different is Canadian society from American society?

I don’t know about the validity of this argument but two sociologists argued a while back that Canada and the United States could be better understand through breaking them into four total regions:

Our research, covering almost 30 years of contemporary and historical analysis, shows the four-regions model fits the evidence much better than a simple two-nations model, in which Canada and the US in general are portrayed as very different. There certainly are other internal differences that could be considered, like those between the US west coast and New England, or between British Columbia and Canada’s Atlantic region.However, we found clear and consistent evidence that the strongest lines of demarcation separate Québec and the rest of Canada, on the one hand, and the American South and non-South, on the other, with national differences usually far less prominent.

In Regions Apart, and in other studies that we and others have conducted, Québec is clearly the most left-liberal region of North America on topics like gay rights, same-sex marriage, common-law marriage, adolescent sexuality, capital punishment, taxation, government spending, unionization, military intervention and so on. The US South is the most conservative or traditional on these same issues. The rest of Canada and the US are usually quite similar on these and other cultural, social, political and economic questions…

What Jim and I called the four “deep structural” principles of the two nations are still intact, though more as ideals to strive for, and not as perfectly achieved realities in either country. These include liberty, individual freedom to pursue one’s goals, while also accepting the rights of others to pursue their goals; equality, the same rights and opportunities for all citizens, though not necessarily the same life outcomes; popular sovereignty, government of the people, by the people and for the people, as Abraham Lincoln so eloquently put it; and pluralism, the belief that all individuals have the fundamental right to be different, even if other people don’t always like or agree with their differences.

As for divergences, I think we have long been divergent in the area of criminal justice, where we see consistently much higher US incarceration and homicide rates, for example. However, even here some differences are exaggerated, for, as shown in Regions Apart, Canada actually has somewhat higher rates for some non-violent crimes, like auto thefts and break-and-entry.

Another area of substantial difference or divergence over the years concerns our roles in the world. The US is far more powerful politically, economically and culturally than Canada, and such differences inevitably give rise to occasionally different views about how to address some of the world’s problems. But we have also been close political allies and economic partners for many decades, so even here our divergent positions can be overstated in many instances, and can regularly change toward more convergence again at a later time.

I don’t know how accurate such an analysis is without looking further at the methodology of how these regions were developed. Why four regions? How was the cluster analysis undertaken? How much variation is within these categories?

At the same time, this made me think: just how much do Americans know about Canada? Could they even identify these two broad regions or some of the key tensions in Canadian life today? On the other hand, I suspect Canadians know more about American life. This could be due to a variety of factors yet it seems odd that we wouldn’t know much about Canada given some of our overlapping background and interests as well as geographic proximity.

Looking at concentrated income in the United States by county

Looking at median household income by county shows some interesting regional patterns in the United States:

There are more than 3,000 counties in the U.S. Of the 75 with the highest incomes, 44 are located in the Northeast, including Maryland and Virginia. The corridor of metropolitan statistical areas that runs from Washington, D.C., through Baltimore, Philadelphia, New York and Boston includes 37 of these top-earning counties (where the median family takes home at least $75,000 a year). Zoom in to the region, and it shows a kind of wealth belt unmatched even on the West Coast.

Poverty is similarly concentrated in the American South. Seventy-nine percent of the poorest counties in the country (where the median family makes less than $35,437) are located in the South..

Relative to 2007, 33 percent of all U.S. counties saw statistically significant increases in poverty by 2012 (across all age groups), deepening the challenges in places that had been struggling even before the recession. Over this same time period, however, one part of the country in particular saw an actual increase in median incomes, and it wasn’t the traditionally wealthy Northeast corridor.

It was the Upper Great Plains. Statistically significant increases in median income, from 2007-2012, are shown in green.

The maps help make these regional patterns clear. But, I wonder how much looking at patterns obscures some important information:

1. Counties are relatively big pieces of land. While income by county tells us something, it also covers up important variation within counties. Take a wealthy county: it doesn’t mean everyone is doing so but just that the median is higher than other places. Think of Manhattan where there are plenty of wealthy people but not everyone there is working on Wall Street or buying luxury condos in new buildings. It would be a lot harder to show on a single map but having 25th and 75th percentile information for each county would help show the relative distributions.

2. These figures aren’t weighted by population. A number of those wealthy Northeast counties have lots of plenty. In fact, perhaps the headline is understated when the population is accounted for. In contrast, the end of the article looks at a few counties where median incomes actually increases – the Great Plains with their new found gas wealth – but there aren’t many people there.

3. It is misleading to have a headline about wealth and talk about wealth in the article when the actual measure being used is median income or poverty levels based on income. Actually, looking at wealth and people’s full assets would likely show even wider gaps between counties.

To reiterate: county-level data can gives us a sense of broad patterns or clusters but may not be the best way to think about income changes in the United States.

Kotkin splits US into 10 areas, predicts which ones will grow

Joel Kotkin takes a look at which areas of the United States will grow in the next decade. Here are a few of his predictions:

Over the next decade, the Left Coast should maintain its momentum, but ultimately it faces a Northeast-like future, with a slowing rate of population growth. High housing prices, particularly in the Bay Area, are transforming it into something of a gated community, largely out of reach to new middle-class families. The density-centric land use policies that have helped drive up Bay Area prices are also increasingly evident in places like Portland and Seattle. The Left Coast has the smallest percentage of residents under 5 outside the Great Lakes and the Northeast, suggesting that a “demographic winter” may arrive there sooner than some might suspect…

The vast region from Texas to Montana has often been written off as “flyover country.” But in the past decade, no nation in America has displayed greater economic dynamism. Since the recession, it has posted the second-fastest job growth rate in the U.S., after the Inland West, and last year it led the country in employment growth. The Dakotas, Nebraska, Oklahoma and Kansas all regularly register among the lowest unemployment rates in the country.

The good times on the Plains are largely due to the new energy boom, which has been driven by a series of major shale finds: the Bakken formation in North Dakota, as well as the Barnett and Permian in Texas. The region’s agricultural sector has also benefited from soaring demand in developing countries…

Once a sleepy, semitropical backwater, the Third Coast, which stretches along the Gulf of Mexico from south Texas to western Florida, has come out of the recession stronger than virtually any other region. Since 2001, its job base has expanded 7%, and it is projected to grow another 18% the coming decade…

More babies and the migration of families, including immigrants, to this low-cost region suggest an even larger political footprint for the Southeast in the decades ahead. Population growth has been more than twice as fast since 2001 as in the Northeast, a trend that is projected continue in the next decade. The region looks set to become smarter, more urban and cosmopolitan, and perhaps a bit less conservative.

Common factors in the analysis: demographics, particularly the influence of immigration; which industries are booming, with an emphasis on technology and oil/gas; and government restrictions/debt. These are common themes for Kotkin. Two quick thoughts:

1. It would be interesting to see how Kotkin’s predictions hold up. One smart move was to restrict this to ten years out which limits some of the unpredictability. But, predictions by experts are notoriously wrong.

2. The prognosis for the South, Southwest, Plains, and some of the West is much better than the big city Midwest or the Northeast. There has been quite a shift to these areas in recent decades in the United States but are we close to a point where these areas take over? Just how much staying power does a region like New York have – would it take decades to overcome its inertia?

Mapping the boundaries of the Midwest

Mapping the Midwest has now become a crowdsourced project through asking “What’s the Midwest to you?”

That’s the question design and planning firm Sasaki Associates is asking visitors to its new exhibit, “Reinvention in the Urban Midwest,” which opens at the Boston Society of Architects (BSA) Space this week. The project includes an interactive survey that contains a timeless challenge: Draw the geographic boundaries of what counts as the U.S. Midwest…

Judging by the maps drawn by others and myself, it appears Montana, Wyoming, Colorado, Kentucky, Arkansas, and Oklahoma are the states of most contention. I personally felt I had no choice but to cut some of them in half. Perhaps the correct answer is still the textbook answer: the states of most intensified yellow (at least as identified by those who’ve lived in the Midwest the longest) make up the U.S. Census Bureau’s definition of the Midwest: Illinois, Indiana, Michigan, Ohio and Wisconsin to the east, plus Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota and South Dakota to the west.  (As a commenter pointed out, cartographer and historian Bill Rankin has also done a Midwest mapping project, in which he overlaid 100 different maps of the Midwest and made the confounding observation that  “no area that was included on every single map”.)

So the geographic boundaries of what most Americans consider the Midwest aren’t exactly clear, but Sasaki has also included another set of maps that reveal a much less murky truth: the Midwest has urbanized in a vastly different way from the rest of the United States. The graphic below maps out the population densities found in urban areas from four U.S. regions in 2010 (a darker shade signifies a larger, denser population)…

The Midwest is characterized by small but strong urban centers that transition sharply to rural surroundings. This pattern has of course grown from the region’s historical focus on agricultural land use. Sasaki’s recent work in Iowa suggests a continued population decline in rural areas but growing population density in more urban areas. However, the growth of urban areas in the Midwest is not uniform. The firm has further identified that agricultural cities in the plains sub-region, such as Des Moines, Iowa, or Lincoln, Nebraska, are indeed growing due to factors like de-ruralization and in-migration to city centers, while traditionally heavy-industry cities in the forest sub-region, such as Milwaukee or St. Louis, are still losing population.

One takeaway: the Midwest is a fuzzily-defined entity that perhaps has more to do with perceptions and culture than it does with exact geography. This would be aided by then asking people who drew the maps to then type in words they associate with the Midwest. I like the contrasting maps between those who have spent more of their lives in the Midwest versus those who have not: there are some clear differences.

The connection between farmland and cities is a good catch. Big cities like Chicago or Omaha were (and still are) intimately connected to agricultural commodities that needed to be distributed and sold through the big cities. For example, if you look at the early railroad construction in the Chicago region, much of it was linked to shipping products from the plains, everything from wheat (southwestern Wisconsin) to lead (Galena) to then distribute further east. Or look at the trading of commodities in places like Chicago and the creation of new kinds of markets. Even though there are big gaps between the Chicago area and the rest of Illinois – they operate as very different worlds – both would strongly consider themselves part of the same region, even if they can’t speak to the deeper ties that connect them.

Kotkin argues a rising South could overtake the North

Joel Kotkin argues that a variety of changes in the South including economic and population growth could help it eventually overtake the North, even with its negative image:

One hundred and fifty years after twin defeats at Gettysburg and Vicksburg destroyed the South’s quest for independence, the region is again on the rise. People and jobs are flowing there, and Northerners are perplexed by the resurgence of America’s home of the ignorant, the obese, the prejudiced and exploited, the religious and the undereducated. Responding to new census data showing the Lone Star State is now home to eight of America’s 15 fastest-growing cities, Gawker asked: “What is it that makes Texas so attractive? Is it the prisons? The racism? The deadly weather? The deadly animals? The deadly crime? The deadly political leadership? The costumed sex fetish conventions? The cannibal necromancers?”…

While the recession was tough on many Southern states, the area’s recovery generally has been stronger than that of Yankeedom: the unemployment rate in the region is now lower than in the West or the Northeast. The Confederacy no longer dominates the list of states with the highest share of people living in poverty; new census measurements (PDF), adjusted for regional cost of living, place the District of Columbia and California first and second. New York now has a higher real poverty rate than Mississippi.

Over the past five decades, the South has also gained in terms of population as Northern states, and more recently California, have lost momentum. Once a major exporter of people to the Union states, today the migration tide flows the other way. The hegira to the sunbelt continues, as last year the region accounted for six of the top eight states attracting domestic migrants—Texas, Florida, North Carolina, Tennessee, South Carolina, and Georgia. Texas and Florida each gained 250,000 net migrants. The top four losers were New York, Illinois, New Jersey, and California…

Bluntly put, if the South can finally shake off the worst parts of its cultural baggage, the region’s eventual ascendancy over the North seems more than likely. High-tech entrepreneurs, movie-makers, and bankers appreciate lower taxes and more sensible regulation, just like manufacturers and energy companies. And people generally prefer affordable homes and family-friendly cities. Throwing in a little Southern hospitality, friendliness, and courtesy can’t hurt either.

Kotkin is determined these days to wave these economic and population figures in the faces of urbanists and coastal residents. That said, the shift to the South is intriguing and significant. Whereas the Northeast and Midwest were ascendent in the early 1900s, the South and Sunbelt have rebounded.

One way this play out with undergraduates is when we discuss how to control for geography in basic regression models. One of the most common ways in sociology to do this is to make a dummy variable for the South. When students ask why this is, I explain that sociologists tend to assume the South is the most unusual region compared to the other three. There are demographic and cultural reasons for this but I wonder if there is some latent feelings about the South…

The home states of military personnel

Richard Florida uses some data to flesh out Defense Secretary Robert Gates recent comment that there is a growing gap between American civilians and the military. Florida suggests part of the issue is the origin of the military personnel: they tend to come from two particular parts of the country.

Aside from relatively high concentrations in Alaska, Hawaii, Washington state, and North Dakota, the military is overwhelmingly concentrated in two distinctive areas of the Sunbelt: The southeast running from Virgina and North Carolina through Kentucky and down through South Carolina, Georgia and Mississippi; and the corridor fromTexas through Oklahoma, New Mexico, Colorado, Kansas and Wyoming. Texas and California now drop out. The upper mid-west and the northeast, especially New England, which tend to be more liberal and left-leaning than the rest of the nation, have very low concentrations of military personnel.

A couple of thoughts:

1. I don’t think this is terribly surprising (though it is helpful to see it in map form).

2. A question: does the military think it might be worthwhile to try to even out these geographic distributions? If so, how could this be done?

3. Are these differences only due to political views (conservatives vs. liberals) or is this really due to social class?

4. I’m glad Florida added data that accounts for differences in population size – the initial map simply showed more military personnel come from more populous states.

h/t Instapundit