This blog is intended to go along with Population: An Introduction to Concepts and Issues, by John R. Weeks, published by Cengage Learning. The latest edition is the 13th (it will be out in January 2020), but this blog is meant to complement any edition of the book by showing the way in which demographic issues are regularly in the news.

You can download an iPhone app for the 13th edition from the App Store (search for Weeks Population).

If you are a user of my textbook and would like to suggest a blog post idea, please email me at: john.weeks@sdsu.edu

Monday, December 26, 2016

The World Really is a Better Place Than it Used to Be

Max Roser at Oxford's Martin School has once again done a magnificent job of pulling together historical demographic data to remind us that there have been tremendous improvements in human existence over the past two hundred years. He says that he did this in response to surveys that have shown that only a tiny fraction of respondents in rich countries think that things are getting better rather than worse. Why don't people know this, he asks? I might answer that not enough people have read my book, and actually that would be a good answer because Roser's data are just providing graphic illustrations of the changes over time that I discuss in detail in my book. Take a look at his charts:


If we think in percentage terms, i.e., a world of 100 people, then we can see that in 1820, 94 of us would have been living in extreme poverty, compared to 10 of today. That is a decline even in absolute terms--from about 940 million in 1820 to 740 million today. Literacy and basic education have improved all over the world, vaccinations are up and child mortality is down. And, in step with the recent blog post by Ben Wilson at LSE about demography and democracy which I noted a few days ago, the percent of the world's population living in a democracy has grown from 1 to 56.

The strength of these numbers is to remind us to be optimistic about the prospect for the continued improvement of human society. The danger, of course, is complacency that might come from thinking that everything is going to automatically be OK. There is nothing automatic about any of this, as Roser notes. It requires massive collaboration. 
There are big problems that remain. None of the above should give us reason to become complacent. On the contrary, it shows us that a lot of work still needs to be done – accomplishing the fastest reduction of poverty is a tremendous achievement, but the fact that 1 out of 10 lives in extreme poverty today is unacceptable. We also must not accept the restrictions of our liberty that remain and that are put in place. And it is also clear that humanity’s impact on the environment is at a level that is not sustainable and is endangering the biosphere and climate on which we depend. We urgently need to reduce our impact.
But then if you've read my book, you already know this... 

Friday, December 23, 2016

The GeoDemographics of Christmas

Younger members of American society (the "Millennials") are more likely to view Christmas as a cultural holiday than as a religious holiday, according to a recent Pew Research poll. That demographic is interesting given the demographic change over time in the Israeli town of Nazareth, where Jesus was from, thus giving rise eventually to the celebration of Christmas (note that according to a Wikipedia entry, Santa Claus first arose from the activities of a Greek Christian Bishop of the 4th Century AD, so both elements of Christmas that are typically celebrated in the US and Europe are linked back to Christianity).

At the time of the birth of Jesus, the population of Nazareth might have been a few thousand, almost all of whom would have been Jewish, although the region at the time was part of the Roman Empire (indeed, you will recall from Luke 2 that the family went from Nazareth to Bethlehem "to be enrolled" in the Roman census). Over time, the city apparently went through transitions in which is was predominantly Christian, but now it has the distinction of being the largest (albeit only about 75,000 people) Arab city in Israel, with about 70 percent of the population being Muslim and the remainder Christian. However, this is a little deceiving because the eastern (old) part of Nazareth was declared a separate city in 1974 and it is predominantly Jewish.

The demography of modern Nazareth brings to mind a link that Abu Daoud sent a few days ago discussing the changing demographics of modern Israel as a whole. The article is titled "Jewish Demography Bodes Well for Israel". The main point of the article is that Israel has the highest birth rate among all developed nations, and that the Jewish population now is reported to have birth rates as high as the Arab population. But the demographics of Israel are still troubling, given the several different groups that comprise the society. A Pew Research report earlier this year noted the following:
Although they live in the same small country and share many traditions, highly religious and secular Jews inhabit largely separate social worlds, with relatively few close friends and little intermarriage outside their own groups. In fact, the survey finds that secular Jews in Israel are more uncomfortable with the notion that a child of theirs might someday marry an ultra-Orthodox Jew than they are with the prospect of their child marrying a Christian. (See Chapter 11 for more information.)
In these increasingly uncertain times, we have to hope that everyone continues to work together in that society as in all societies, regardless of religion or religiosity.

Thursday, December 22, 2016

Population Growth Slows in US Generally; But Not in Utah

The U.S. Census Bureau reported this week on the growth of the population between 1 July 2015 and 1 July 2016. The fastest rates of growth by state were in Utah, Nevada, and Idaho. Utah has had the highest birth rate in the U.S. for many years, fueled by its large (and obviously growing) Mormon population. Idaho also has an increasing Mormon population. We're not sure about Nevada because, as you know, what happens there stays there...

In overall numbers, Texas had the highest numerical increase--an increase of more than 400,000 people in 2016 compared to the previous year. California remains the most populous state in the union, but were it not for continued immigration, the population would probably stop growing due to internal migration out of California to other states--like Texas where property taxes and business taxes and regulations are lower.

Regionally, population growth was highest in the West and the South, as noted by the NYTimes coverage of the Census Bureau's report:
“The movement to the South and West is a very long-term trend,” said Mr. [Jeff] Passel of Pew [Research], adding that those regions attract older residents of the Northeast and Midwest looking for more temperate places to retire.
The Midwest expanded by nearly 0.2 percent, while the population in the Northeast remained virtually unchanged. Both regions lost more residents than they gained from migration, though that was offset by more births than deaths.
The South is now home to 38 percent of the national population, while 24 percent of Americans live in the West. The Midwest is home to 21 percent of the population and the Northeast is home to 17 percent.
The NYTimes also notes that the rate of population growth in the country as a whole is at its lowest level in many decades. Nonetheless, we still added more than 2 million to the total last year, since we are building on a base that represents the third most populous country in the world. And, since we have the largest economy in the world, those new folks add disproportionately to the environmental consequence of ever-rising resource utilization. I was reminded of this by another story this morning about the newest alarm bells going off because of the rise in temperature in the Arctic.

Tuesday, December 20, 2016

Demography and Democracy

Ben Wilson at LSE has posted a very nice blog in Demotrends about the relationship between demography and democracy, building on his work with Tim Dyson. Their analysis is that the kinds of societal changes associated with the demographic transition are those that also tend to promote democracy in societies:
In the second half of the demographic transition, the proportion of adults in the population becomes larger. In demographic terms, fewer children means that the number of people eligible to vote increases, as does overall civic engagement. Put bluntly, children aren’t interested or engaged in politics. 
Another consequence of demographic transition is that when women are caring for fewer children they are more likely to enter the labour market. Women are able to pursue careers, skills, and training in a way that they aren’t able to if they are caring for large families. As a result, they tend to have a greater interest in the political system and their rights.
They then show the statistical relationship between the Vanhanen Index of democracy and the median age as a proxy for where a country is in terms of the demographic transition. The relationship isn't perfect, but it is very intriguing. Now, full disclosure, I have been following Tim Dyson's work forever, and I doubt that I have ever disagreed with anything he wrote. Indeed, if you go back and look at my blog post about the Colombia peace agreement, you will see some similarities with this theme that democracy and demography are intimately related.

Monday, December 19, 2016

Migrants, Migrants, Everywhere

Yesterday was International Migrants Day. I celebrated it by being out of town, but I didn't personally do any migrating. However, a lot of people do migrate all the time, and the Humanitarian Information Unit of the U.S. State Department has put together a very nice map, using UN data, to illustrate the truly global nature of international migration. 


As you look at this infographic, a couple of points they make stand out. First of all, "only" 244 million of today's 7.4 billion people are international migrants. Although it is a big number, it is actually a pretty small fraction of the total, given the amount of fuss and bother that is caused by migration. Secondly, the majority of migrants are relocating to places within their own region, so the spatial scope of international migration is also more limited than it might seem from the press coverage it receives. 

Of course, the issues surrounding migration are what make even proportionately small numbers seem big. No matter where people migrate, they are apt to be perceived as being different in some way that will lead to discrimination. Furthermore, some of those immigrants will wind up not being very integrated into their host society and that will exacerbate local xenophobic tendencies. This is always made more difficult when the immigrants are refugees or asylum seekers (25 million, as you can see), meaning that they have left a lot of their life (and probably loved ones) behind, perhaps never to return. The situation in Aleppo cannot do anything but break your heart, for example.

Friday, December 16, 2016

Central American Asylum Seekers on the Rise

Life in Central America--especially in Guatemala, El Salvador, and Honduras--can be downright dangerous for a lot of people, and the natural human response is to get out. This affects both Mexico and the United States. Reuters reports that the Mexican Commission for Refugee Assistance (COMAR) is anticipating a continued rise in the number of Central Americans seeking asylum.
Mexico expects a sharp increase in people seeking asylum from Central America next year, fleeing gang warfare and poverty in their home countries, a senior official said on Thursday.

There has been a steady surge of Central Americans applying for asylum in Mexico since 2015. Cinthia Perez, a director of Mexico's refugee agency COMAR, said in an interview that she is receiving about 9 percent more applications each month.

There were 3,424 asylum applications in 2015, and she predicts ending 2016 with around 8,000. That figure could well rise to 22,501 by the end of 2017 if the trend of 9 percent more applications each month continues.
Many of these people will stay in Mexico, but once granted asylum they have freedom of movement within Mexico, and who knows how many will head to the U.S.-Mexico border. The election of Donald Trump seems to have spurred an increase towards the border in anticipation of potentially even tighter controls after Trump assumes office.

A related story from Prospect.org this week offers some background into the violence in Central America, and the role played by U.S. deportation policies in that violence. There is also a link there to a story last year in the Guardian in which my PhD student, Elizabeth Kennedy, was quoted regarding her very important research on this issue.

Wednesday, December 14, 2016

Xenophobia Rules! Overestimates of the Muslim Population in Western Countries

Fear of an influx of Muslim immigrants seemed to fuel at least some voters in the U.S. to vote for Donald Trump, and polls in the U.K. suggested that people with the least contact with immigrants were the ones most likely to vote to leave the EU, as I noted at the time of that vote. Now a new set of poll data from western countries shows the depth and scope of the issue: People tend to vastly overestimate the number of Muslims in their own country. The Guardian covers the story, emphasizing that the reality gap is widest in France.
An Ipsos Mori survey that measured the gap between public perception and reality in 40 countries in 2016 found French respondents were by far the most likely to overstate their country’s current and projected Muslim population.
The average French estimate was that 31% of the population was Muslim – almost one in three residents. According to Pew Research, France’s Muslim population actually stood at 7.5% in 2010, or one in 13 people.
The French were not the only ones to hold such misconceptions: Italian, German and Belgian respondents all guessed that more than a fifth of the resident population was Muslim, while in reality the figure ranges from 3.7% in Italy to 7% in Belgium. All three countries also greatly overstated the expected proportion of Muslim residents in 2020.
The graph below shows the results for several countries, including the U.S. You can readily imagine that the discussions in the media about refugees--especially inspired by the crisis in Syria--have increased and distorted the public perception about the presence of Muslims in western countries. 


Tuesday, December 13, 2016

Refugee Adaptation Depends in Part on Their Neighbors

In my view, the single biggest concern that host societies have about incoming refugees is: How will they adapt to the new society? Will they eventually integrate and fit in, or will they stay in an ethnic neighborhood that essentially preserves the old culture and protects them from the new one. How much should governmental influence these decisions? A new infographic from Population Europe highlights research that puts an interesting spin on this issue, based on Sweden's experiences with Turkish and Iranian immigrants. In a paper accepted for publication in Population, Space, and Place, the authors use a longitudinal database to track resettlement patterns. They find that the region of origin (the place within a country) is more important in resettlement decisions that is the country per se.
Imagine Kurds from Turkey or Iran who thanks to home-country conflicts prefer to live alongside other Kurds but not Turks or Persians. Thus, ethnic conflicts in the sending country may manifest themselves in immigrants’ residential choices in the host country. Whether or not to encourage relocation, the drivers of post-migration residence decisions are key to good asylum policies. Otherwise, carefully crafted common asylum systems could be swiftly undermined.
In other words, ethnic differences are at play not only in the residential preferences and adaptation of immigrants to the host society, but also in the way that immigrants themselves behave toward each other in the host society. 

Monday, December 12, 2016

Why is Germany's Birth Rate so Low?

Germany's birth rate has been below replacement level since the early 1970s, according to data from UN demographers. There have been many lamentations about this fact in Germany, but until just about this year there were enough women of childbearing age--enhanced by immigration, especially from Turkey--having children that the population did not decline. It is threat of population decline, along with the not-so-popular solution of letting in more immigrants, that is one of the forces behind the right-wing populist movement in that country. So, I was very surprised by an article in OZY yesterday discussing results from a YouGov poll suggesting a high level of regret among Germans about becoming parents.
YouGov undertook what’s believed to be the first quantitative “parenthood regret” survey earlier this year, quizzing 1,228 parents — 671 women and 557 men. It found that a whopping 19 percent of mothers and 20 percent of fathers regret having children. They love their offspring but, if given the choice, would steer clear of parenthood a second time around.
The reasons surrounding this regret, at least for women, have to do with the lack of accommodation in German society for combining parenthood and a career. 
Young German women often struggle to find permanent employment following their studies, says Geissler, and the country’s generous maternity leave — three years with one’s post reserved — is available only to those who have such job stability. There also are limited options for working part-time as a professional. Nele Dagefoerde, a business-degree-toting mother of four in Esslingen, just outside Stuttgart, says that as long as companies and society fail to provide convincing models for moms to return to work part-time, and make it easier for men to do the same, it will always be difficult to have both a career and a family. “As long as women still have to make a choice whether to go for a career or to become a mother, there will be moms who regret motherhood,” she says.
There also appear to be limits to child-care combined with schools that end at mid-day that are oriented around a more traditional world of non-working mothers. Reading between the lines, you can see that Germany (like Japan, as well, as I've noted before) cannot reconcile itself to a world in which women are encouraged to become educated and employed in the paid labor force, and yet they are also expected to be full-time mothers. Women are essentially forced to choose and a lot of them decide to forego children as a consequence. Germans have known about this for several decades, yet have clearly done very little about it.

Sunday, December 11, 2016

Money and Age Still Go Together in the U.S.

As I discuss in the essay in Chapter 10 of my text, the wealth in the United States is concentrated in the population aged 65 and older. We see this every year when Forbes publishes its list of the 400 richest Americans. Now, to be honest, they put out this year's list several weeks ago, but it took me a while to organize the data into a spreadsheet so that I could analyze it (and, yes, you can pay Forbes for a spreadsheet, but I declined that offer) for an update to what's in the Chapter 10 essay in the 12th edition. The bottom line: in 2016 people 65 and older account for 57% of the richest 400, and they hold 59% of the wealth in that group.

At the same time, we have to recognize that 3 of the 5 richest Americans are under age 65: Bill Gates is 61, Jeff Bezos is 52, and Mark Zuckerberg is 32. If you are rich at a young age, you are likely to be uber-rich. Keeping in mind that it took a net worth of 1.7 billion to get in this club, people less than 45 years old had an average wealth of 9.8 billion, compared to 6.1 for those aged 45-54, 4.8 for those aged 55-64, and 6.2 for those aged 65 and older.

Although the list shows that there are a lot of ways to get rich (James LePrino has made billions out of Mozzarella cheese, for example), the one that stands out in terms of how many people are on this list is finance--in some way or another. That is, of course, the source of wealth for Warren Buffet--the third richest American, but there are a lot of hedge fund, private equity, and other investment-related people on the list. If you are watching "Medici" on Netflix, as my wife and I are, you will appreciate the historical nature of finance as a way of "financing" wealth. And if you want some insight to how private equity firms enrich their partners, check out this article in today's NYTimes.

Friday, December 9, 2016

Life Expectancy at Birth Takes a Hit in the US

Life expectancy in the United States is already the lowest among the rich nations of the world, despite the highest per person cost in the world. More bad news came out this week from the U.S. National Center for Health Statistics (NCHS--part of the Centers for Disease Control): life expectancy at birth dropped between 2014 and 2015. To be fair, the drop was minuscule, but a decline is still going in the wrong direction. The media, such as the Washington Post, emphasized that this was the first drop since 1993. I admit that I didn't remember that drop, but a little investigation revealed that it was due to deaths from HIV reaching their peak among men aged 25-44.

The latest increase appears to have multiple causes. First, we have the increase in opioid-related deaths, already identified by Anne Case and Angus Deaton at Princeton. Then there is the continuing increase in obesity and diabetes in the United States that is especially troublesome prior to old age. This is noteworthy because the report from the NCHS shows that life expectancy held steady at ages 65 and older. It is the younger ages that are driving the decline. And, on that note, we need a hat tip to Beth Jarosz, who tweeted out that the infant mortality rose between 2014 and 2015, although the media tended to ignore that bit of important information.

I saw today that the most-read article this year from "The Nation's Health", published online by the American Public Health Association, was the story about where U.S. states rank in terms of health. That story, while generally positive about health in the U.S., also pointed out the dilemma of obesity and diabetes. Put another way, we need a medical model that emphasizes prevention, rather than just expensive treatment. So, next time someone asks if we need health care reform, the answer is yes, and in a big way.


Wednesday, December 7, 2016

Abortion Rights Under Attack Yet Again--UPDATED

As I've said before, I am 100% convinced that if both men and women could get pregnant, abortion would be legal everywhere. But that's not how we evolved, so we have to continually cope with attacks on women's reproductive rights. The latest is a law just passed in Ohio, as reported by CNN:
Donald Trump's election, and a presumption that he'll appoint conservative Supreme Court justices, spurred Ohio Republicans to pass what would effectively be the nation's strictest time-based abortion law, a legislator said. 
Ohio lawmakers on Tuesday passed a controversial "Heartbeat Bill" that would ban abortions in that state from the moment the heartbeat of a fetus can be detected -- which usually occurs about six weeks into a pregnancy.
Two things strike me about this. First, as noted on NPR's reporting about this, most women don't even know they're pregnant at 6 weeks, so this legislation is effectively taking that option away from them. Secondly, of course, is the idea that Trump's election will dramatically increase the pressure on abortion rights. This is the same pressure being felt by people like Dr. Warren Hern in Colorado, whom I discussed shortly after the election.

As I write this, Ohio governor John Kasich has not voiced an opinion about the bill, but everything that he has said in the past suggests that he would be sympathetic. He has another week to veto it, but if he does nothing, it automatically goes into law next year.

UPDATE (12/13/16): Governor Kasich did, in fact, veto this bill, although he signed into law a bill outlawing abortion after 20 weeks in Ohio. It is possible that the more extreme bill was just a smoke screen to make it easier to pass the 20 week law, but I am just guessing on that.

Tuesday, December 6, 2016

Cultural Differences in Fertility Levels in the U.K.

The United Kingdom has been generally welcoming to immigrants from its former colonies in Africa, Asia, and the Caribbean. Each group has brought with it various cultural differences, including religion, that shape the U.K., especially in urban areas where most immigrants settle. The expectation in demography generally is that immigrants will have higher fertility than the native population, but that their children will reproduce at a rate closer to that of the native population. Hill Kulu and Tina Hannemann have found an exception to this pattern in the U.K. among descendants of immigrants from Pakistan and Bangladesh. Their findings just came out in the online journal Demographic Research. Here is what the data show, drawing upon the longitudinal Understanding Society project in the U.K.:


Kulu and Hannemann show that Pakistani and Bangladeshi women are just as likely to have a first child at about the same age as natives. So, the starting of a family is not different. What differs is that they are much more likely to have a second, third, and fourth child. This is not attributable to differences in education or employment status. Rather, the difference seems to be largely a function of culture, which includes religion (almost all immigrants from those two countries are Muslim), and coming themselves from and valuing large families. 

Of particular interest, though, is the finding that there is a certain bipolar distribution among immigrant groups and their descendants. Not all women act alike. Some (albeit still the minority) delay marriage and childbearing, while others are more likely to be the ones having a fourth child. The future will vary considerably depending upon which group gains the most traction over time.


Monday, December 5, 2016

Is the World's Muslim Population Growing Faster than the Christian Population?

We could perhaps call it the Religion Race. The two great proselytizing religions in the world are Christianity and Islam. Christianity had a seven century head start on Islam, but Islam has been steadily catching up with Christianity in terms of the overall number of adherents. Thanks to Abu Daoud for pointing me to a recent assessment of where we stand globally.
Bottom line: both Christianity and Islam are growing faster than the world population (so the world is becoming more religious); and Islam is growing faster than Christianity, so the % Muslim in the world is increasing.
These numbers are based on projections made by the Center for the Study of Global Christianity at the Gordon-Conwell Theological Seminary in Massachusetts. Although they use UN Population Division projections for their baseline numbers, they clearly have to make their own estimates of the population by religion. Their data project that by 2050 there will be 3.4 billion Christians, accounting for 35% of the world's projected population of 9.6 billion. By comparison, they project a total of 2.7 Muslims in 2050, representing 28% of the world's total. 

Meanwhile, last year Pew Research published its projections of the world population by religious affiliation. They projected a total of 9.3 billion people in the world by 2050, of whom 2.9 billion are projected to be Christian (31%), and 2.8 billion Muslims (30%). So, we see that the Christian group making projections has a higher number for members of its religious group than does the secular research organization. Does that surprise us? I think not. In either case, the conclusion stands that largely through birthrate differences, the Muslim population of the world is growing faster than the Christian population.

Sunday, December 4, 2016

Defining U.S. Regions Using Census Data on Commute to Work

Humans have a natural sense of place and we tend to define ourselves at least partly by where we live. But how do we define the boundaries of places? Political entities are typically used and they may or may not reflect the lives of everyday people. Two researchers have nicely utilized commute to work data from the American Community Survey to organize the U.S. into what might be called "organic" (as opposed to politically defined) regions. The results are published in the open source journal PLOS ONE and summarized on the Atlasobsura.com website. The first map below summarizes the commuting patterns between census tracts in the U.S.:

Now, to be sure, some of those commutes are small in number, but that is all discussed in the paper (which you should read before making a final decision about how you feel about this). Now, combining these linkages with visual interpretation, the researchers (Garrett Dash Nelson, a historical geographer from Dartmouth, and Alasdair Rae, an urban analyst from the University of Sheffield) came up with the following map of "real" regions of the U.S.:

Since the baseline data refer to commuting patterns, these regions represent what we might think of as "spheres of influence" of major cities. Thus, you have one big area in the western states that lacks big cities and thus lacks any real commute patterns. Maybe that's what's left of the "wild west."

Friday, December 2, 2016

Why are People Left Behind in Modern Society?

The general sentiment that I have seen regarding Donald Trump's election is that many people in America are feeling left behind and felt as though nobody in the current government was doing anything about that. Part of the explanation for people being left behind is, indeed, because many manufacturing jobs have been sent off to developing countries where wages are much lower and where, in general, environmental regulations and other worker and consumer protections are not built into the cost of manufacturing. The result is actually good for people because the increased availability of affordable consumer goods in this and other countries essentially raises everyone's standard of living since things are cheaper to purchase than they used to be. The availability of these cheap imported goods is the secret behind the amazing success of the Walton family--Wal-Mart opened its first store in 1962, shortly before the globalization of the labor market really took off. 

At the same time, however, manufacturing actually continued unabated in the U.S., but automation is replacing workers, as demonstrated by a Washington Post article intriguingly headlined "A single chart everybody needs to look at before Trump's big fight over bringing back American jobs..." The graph is from a report by researchers at the Brookings Institution and makes the point that "U.S. factories now manufacture twice as much as they did in 1984, with one-third fewer workers." Here's the graph:


The workplace is different than it used to be and people (including society in general) have to respond. But it is hard when there are obstacles to social mobility and employment success. The biggest obstacle, according to two new studies (adding fuel to a long-running hypothesis) is the family into which a child is born. A report out from the National Academy of Sciences concludes that "Parenting Matters."
From birth, children are learning and rely on parents and the other caregivers in their lives to protect and care for them. The impact of parents may never be greater than during the earliest years of life, when a child’s brain is rapidly developing and when nearly all of her or his experiences are created and shaped by parents and the family environment.
And a study out of Stanford University's Center on Poverty and Inequality shows that the U.S. and U.K. lag behind other culturally similar countries in ways that limit a child's chances of social mobility. The Financial Times summarizes the situation as follows:
Median family income was higher for rich families and lower for poor ones in the US and the UK than in Australia and Canada. In low-income families, the proportion where the mother was in poor health was far higher in the UK and US. 
The percentage of children born to a teenage mother was strikingly higher in poorer families in the US and significantly higher in the UK than in Canada or Australia.
In other words, we start out with too much income inequality (exactly as Thomas Picketty has famously talked about), but then compound that with high levels of out-of-wedlock births which lead to limited means available for children's development, not to mention the limitations that this places on the mother herself. 

Thursday, December 1, 2016

Many of China's Missing Girls May Not be Missing After All

A widely researched aspect of China's one-child policy, implemented back in 1979, was the rise in the sex ratio at birth. Census data revealed that there were far fewer girls at the youngest ages than you would have otherwise expected. The likely explanation seemed to be that millions of female fetuses had been aborted, or even that female infanticide was being practiced. While some of that probably did happen, a paper just published in the journal China Quarterly, suggests that many of these girls were in fact born, but were not registered until later--often years later. This was made possible largely through the complicity of local officials, largely in rural areas of China. The Washington Post has a summary, quoting John Kennedy at the University of Kansas, one of the study's authors:
“Most people are using a demographic explanation to say that abortion or infanticide are the reasons they don't show up in the census and that they don't exist. But we find there is a political explanation.” Local officials, they argue, were complicit in the concealment to retain support from villagers, and maintain social stability. “There is no coordination between cadres saying 'we're all in agreement,'” Kennedy said. “Actually it's just very local. The people who are implementing these policies work for the government in a sense. They are officials, but they are also villagers, and they have to live in the village where they are implementing policies.”
The authors first formed this idea back in 1996, but since one of the authors (Yaojiang Shi) is at a university in China, it was politically too risky until very recently to broach the subject publicly. The authors analyze age cohort data for older ages to show that, after accounting for probabilities of death, the sex ratios at older ages are closer than would be expected based on the highly skewed sex ratios at birth based on registered births. Parents were eventually paying the fine for the child so that she could attend school and eventually marry. Without her hukou registration, these things would not be possible.

These practices shed sad light on gender inequality in China. As the authors note in their paper: "This is associated with the virilocal marriage system whereby girls are raised by their natal family but then live with their husband’s family after marriage. Traditionally, daughters are considered to be 'born into another’s family.' As a result, there is no social or economic incentive for families, especially in the countryside, to have daughters." On the other hand, the acknowledgement that many of these missing girls aren't actually missing is good news for a government that has effectively ended the one-child policy in order to boost the age structure at the younger ages. And it means that the marriage for Chinese men may not be as bad as had been perceived, at least not in rural areas. 
                       

Tuesday, November 29, 2016

Political Demography and Latin America

Today I was the guest on Understanding Latin American Politics: The Podcast, which is hosted by Dr. Gregory Weeks, Professor and Chair of Political Science at the University of Carolina, Charlotte. Greg has always been like a son to me since, of course, he IS my son. Not surprisingly, we talked about political demography, which he and I have been working on together for several years. Today we talked especially about Colombia, Venezuela and Cuba. Enjoy!

Monday, November 28, 2016

Is Demography Connected to the Colombian Peace Deal?

A new peace deal was signed in Colombia a few days ago between the government and the FARC rebel group. As BBC News notes, "The deal is aimed at ending five decades of armed conflict, which has killed more than 260,000 people and left millions internally displaced." That's a lot of deaths, to be sure, but 50 years ago Colombia had a population of 19 million, and today it has 48 million. Fifty years ago, the average woman was having 6.2 children each and today she is having 1.8 each. Fifty years ago, almost 60 percent of the population was under age 20, and today it is down to one-third. In other words, Colombia is a very different country today than it was 50 years ago, and demographic change is front and center in the country's transition.

I thought about these things this week as I read an interesting research article just published in Demographic Research by Ewa Batyra, a demography PhD student at the London School of Economics. Her focus was on the change in timing of childbearing in Colombia as a way of helping to understand the dramatic drop in fertility over time in that country. Her analysis is largely based on the 2010 Demographic and Health Survey in Colombia, at which time fertility had dropped to just replacement level. It has since dipped below replacement level. Why? As is true throughout much of the world, education is a key. The better educated a woman is, the later she is to start childbearing or, if she had a child early (as is still frustratingly common throughout parts of Latin America), the more likely she is to delay the second one.

So, in the 50 years since FARC got going, the excess supply of unemployed young men has diminished, and men and women have been staying in school longer, leading to new perspectives on life, including perspectives on family and, almost certainly, on the value of rebellion and illegal activity. Over those years, the per person gross national income (in current US dollars) has risen from $300 per year to $7,130, according to World Bank data. 

You might well say that correlation is not causation and that I am ignoring a lot of the messiness associated with the peace deal, but I would argue that Colombia's way forward (and it is definitely forward) is grounded thoroughly in the demographics of reduced family size and a heightened status of women. The need for rebellion is gone, and it was time to make peace. 

Saturday, November 26, 2016

Venezuelans on the Move

It was bound to happen. You can't have a country whose economy is collapsing, as in Venezuela, and not expect that some people will try to get out. Today's NYTimes has a lengthy article detailing the plight of several people trying to get to the island of Curaçao by boat. From shore to shore it is about 40 miles in distance, but of course that's not an easy trip for small smuggling boats. And the island is home to only about 150,000 people, so it is not in a good position to be receiving new migrants.
Venezuela was once one of Latin America’s richest countries, flush with oil wealth that attracted immigrants from places as varied as Europe and the Middle East.
But after President Hugo Chávez vowed to break the country’s economic elite and redistribute wealth to the poor, the rich and middle class fled to more welcoming countries in droves, creating what demographers describe as Venezuela’s first diaspora.
Now a second diaspora is underway — much less wealthy and not nearly as welcome.
Well over 150,000 Venezuelans have fled the country in the last year alone, the highest in more than a decade, according to scholars studying the exodus.
I have not been able yet to track down the source of those numbers, but the story points out that Venezuelans are also trying to get into Brazil (whose economic situation has also deteriorated lately), as well as into Colombia. 

A cynical view of this would be that President Maduro of Venezuela might be happy to see people go, since that could take some pressure off the limited supply of food in the country.

Wednesday, November 23, 2016

Federal Court Pushes Back Against Gerrymandering in Wisconsin

Federal judges in Wisconsin have ruled that the gerrymandering of State Assembly districts in that state unfairly benefitted Republicans. Keep in mind that redistricting of Congressional Districts is mandated by the U.S. Constitution and this is the formal reason for having a census of population in the United States. State legislatures and local jurisdictions also redistrict after each census. Indeed, my wife (a former elected official) was a member of the San Diego County Redistricting Advisory Committee here in San Diego County after the 2010 census. The Supreme Court has ruled in recent years that race is the primary reason for challenging the redrawn boundaries, but in Wisconsin, political parties were at stake, rather than race, per se, as the NYTimes notes:
A panel of three federal judges said on Monday that the Wisconsin Legislature’s 2011 redrawing of State Assembly districts to favor Republicans was an unconstitutional partisan gerrymander, the first such ruling in three decades of pitched legal battles over the issue.
Federal courts have struck down gerrymanders on racial grounds, but not on grounds that they unfairly give advantage to a political party — the more common form of gerrymandering. The case could now go directly to the Supreme Court, where its fate may rest with a single justice, Anthony M. Kennedy, who has expressed a willingness to strike down partisan gerrymanders but has yet to accept a rationale for it.
In Monday’s ruling, the court was swayed by a new and simple mathematical formula to measure the extent of partisan gerrymandering, called the efficiency gap. The formula divides the difference between the two parties’ “wasted votes” — votes beyond those needed by a winning side, and votes cast by a losing side — by the total number of votes cast. When both parties waste the same number of votes, the result is zero — an ideal solution. But as a winning party wastes fewer and fewer votes than its opponent, its score rises.
The case will automatically go to the Supreme Court and we will have to watch this carefully, since gerrymandering for clearly political purposes was undertaken by many states after the 2010 census because Republicans won a lot of seats that year both in Congress and in state legislatures. Since the latter groups still draw the boundaries in a majority of states, this has almost certainly influenced politics in America, as things suggested they would back in 2011 as redistricting was taking place throughout the country. If boundaries are required to change in ways so that one party is clearly not benefitting more than another, this could affect elections in 2018 and 2020, which could then influence how the 2020 census data will be used for redistricting.

Tuesday, November 22, 2016

The Demography of Dementia

With aging comes dementia. Maybe not to you personally (hopefully!) but at the population level, your chances of dementia increase with age. But how quickly do those odds increase? A new report out in JAMA Internal Medicine offers a bit of good news, if there is such a thing when it comes to dementia. You can read the whole article on your own, but today's NYTimes offers a summary.
Despite fears that dementia rates were going to explode as the population grows older and fatter, and has more diabetesand high blood pressure, a large nationally representative survey has found the reverse. Dementia is actually on the wane. And when people do get dementia, they get it at older and older ages.
The new study found that the dementia rate in Americans 65 and older fell by 24 percent over 12 years, to 8.8 percent in 2012 from 11.6 percent in 2000. That trend is “statistically significant and impressive,” said Samuel Preston, a demographer at the University of Pennsylvania [and a Past President of the Population Association of America] who was not associated with the study.
In 2000, people received a diagnosis of dementia at an average age of 80.7; in 2012, the average age was 82.4.
Keep in mind that this is a good new/bad news sort of story. The good news is that the rate of dementia is going down and is hitting at later ages. The bad news is that given the size of the baby boomer population coming into the older ages, that rate is going to have to go down a lot more to ward off a huge national increase in the number of older people with dementia. Nature had a lengthy story recently that details some of those numbers and the fact that we have not yet put nearly enough money or effort into preventing and combatting dementia.
Voluntarily or not, people will need to face up to dementia, because in just a few short decades, pretty much everyone is going to have a friend or loved one affected by the disease. It’s an alarming idea, and it should spur action, says Robert Egge, chief public policy officer of the Alzheimer’s Association in Chicago, Illinois.
So, we can't let the good news from today's research keep us from pressing on with more work on dementia.  

Monday, November 21, 2016

Pope Extends Forgiveness for Abortions

Pope Francis today extended indefinitely his instruction to parish priests that they can forgive a woman who confesses to having had an abortion. As CNN notes, this is not meant by the Catholic Church to be a change in their view about abortion, but it does show a continued softening (even if still somewhat confused, as I have noted in the past) of attitudes towards women's reproductive rights that has been taking place under the current Pope.
While the practical effect of Francis' announcement remains unclear, it draws attention to the prevailing theme of his papacy: That the doors of the Church must remain open, just as God's forgiveness and mercy extend to all those who repent from sin. 
That said, the Catholic Church's stance on abortion has not changed -- it is still viewed as a "grave sin." But it makes it easier for women who have had abortions to be absolved for their actions, and rejoin the church.
The CNN article has a nice map showing the availability of legal abortion throughout the world (note that the original is interactive, allowing you to to click on countries to find their specific laws regarding abortion):


The map shows a significant "north-south" divide with respect to abortion. Interestingly enough, the lines are drawn more with respect to levels of economic development (more developed equals greater availability of abortion) than along religious lines (some predominantly Catholic countries allow it, some don't; some predominantly Muslim countries allow, some don't). 

My guess is that if men could get pregnant, abortion would be available everywhere in the world...

Sunday, November 20, 2016

The Great Ethnic Wall of China

"Ethnicity is central to China's national identity," reports the Economist in this week's issue. The Han account for 1.2 billion of the country's 1.4 billion people, and it seems clear that they want it to stay that way.
China today is extraordinarily homogenous. It sustains that by remaining almost entirely closed to new entrants except by birth. Unless someone is the child of a Chinese national, no matter how long they live there, how much money they make or tax they pay, it is virtually impossible to become a citizen. Someone who marries a Chinese person can theoretically gain citizenship; in practice few do. As a result, the most populous nation on Earth has only 1,448 naturalised Chinese in total, according to the 2010 census. Even Japan, better known for hostility to immigration, naturalises around 10,000 new citizens each year; in America the figure is some 700,000.
This issue is at work behind the scenes, so to speak, as China copes with its changing age structure. The dramatic drop in China's fertility rate from the 1970s to the present is a key factor in its economic success. The country has used that demographic dividend brilliantly--as have Taiwan, Japan, South Korea, and to a less extent, Vietnam. As I note in my book, the evidence suggests that the birth rate decline in China would have continued even without the Draconian one-child policy, and the negative legacy of that policy lingers even though the government has eased the restrictions on births. The NYTimes recently reported that a 2015 mini-census in China found that the Total Fertility Rate in China was down to 1.05 children per woman, well below the official figure of 1.6 children. The response from some has predictably been to suggest that the government needs to be more pronatalist. The government has shown little interest in this approach, nor has the average married couple in China. And, since it is clear that outsiders (i.e., immigrants) are not going to be coming to the rescue anytime soon, we can assume that China will continue its pattern of creating a global web of external income that will allow it to support an increasingly older population without resorting to a higher birth rate among the Han or opening the doors to non-Chinese immigrants.

Friday, November 18, 2016

Afghans on the Move

Most of the news out of Afghanistan is about Taliban-inspired violence. In the midst of the chaos, however, are real people trying to stay out of harm's way. That means that a lot of Afghans are on the move. I mentioned this a few days ago, albeit largely from the Pakistani point of view. The Humanitarian Information Unit of the U.S. State Department has just published an infographic with lots of useful details from the Afghan point of view. I have copied it below, but you should go to the original to see it more clearly.

Afghanistan has an estimated 33 million people, and although they are sandwiched in between Pakistan and Iran, the demographics look much more like Africa than either South Asia (in which both Pakistan and Afghanistan are located) and West Asia (where geographers put Iran). For a long time, Afghanistan has had high fertility, high infant and child mortality, and high maternal mortality. A quick glance at the age structure of returnees (which is not unlike the age structure for all of Afghanistan) tells you how disruptive the future demographics are likely to be. 

Thursday, November 17, 2016

Disruptive Demographics

The term "disruptive" has come into wide use (and maybe even overuse) in the past few years, but it does apply to the effects of demographic change. I talk about the effects of demographic change all the time, including recently in the context of political demography, but I had not before used the term "disruptive." An article in The Economist's online section from the Economist Intelligence Unit taught me that I need to make it mine. 
Demographic forces are disruption catalysts. Young people’s online habits have transformed advertising and customer feedback models. Older generations’ purchasing decisions are increasingly influenced by their children, quickening the percolation of new technologies. And in some new frontier technologies, like robotics-in-the-home, the ‘silver market’ may be the first adopters.
The story is about marketing (which I discuss in my book), not about politics, but the disruptive force of demography is even more powerful with respect to politics, as Debbie Fugate and I talk about in our book on The Youth Bulge: Challenge or Opportunity. If we were doing that book this year the title would have to be The Demographic Disruptions of Youth Bulges.

Wednesday, November 16, 2016

U.S. Women Are Marrying Less and Divorcing Less

Divorce and contraception (and abortion) became widely available to U.S. women at roughly the same time back in the 1970s. At first there was a rise in the divorce rate, but over time it is the decline in the marriage rate among women that has been most noticeable. These trends have been put out there for us by the folks at Bowling Green State University's National Center for Marriage and Family Research. Here's what the data look like over time:


You can see the long-term decline in the marriage rate among women, with a slight rise in the past few years. At the same time, the divorce rate went up in the 1970s, but then leveled off and has dipped a bit in the last decade. These are crude rates--not adjusted for age--so they should be interpreted with a bit of caution, but it seems clear that women have been making the decision not to marry (or to delay getting married--we know that is happening), thus reducing the risk of divorce. Note also that the U.S. government stopped gathering marriage and divorce data several years ago, so this information now comes from questions asked on the American Community Survey. You may recall that the Census Bureau talked about dropping those questions, but people like us vehemently objected and so they didn't.

Tuesday, November 15, 2016

Undocumented Immigrants Being Forced Out of Pakistan

Even as Donald Trump continues to threaten to deport undocumented immigrants currently living in the U.S., Pakistan is threatening the same to several hundreds of thousands of Afghans who have been living in Pakistan without documents. Some of these people have been in Pakistan since the Soviet Union invaded Afghanistan back in 1979. CNN notes that there is a mix of refugees and migrants, but they all share the same situation of lacking proper documentation.
Currently, all undocumented Afghan migrants and refugees in Pakistan face an official deadline of November 15 to secure documentation or leave. But pressure has been growing for the authorities to extend the deadline to allow more time for the returnees to transition peacefully back to Afghanistan.
So far this year, 210,998 undocumented migrants and refugees have returned to Afghanistan, with numbers soaring in recent months, according to the International Organization for Migration. On top of that, there are also hundreds of thousands of registered Afghan refugees living in Pakistan, who face a deadline of March 31, 2017 to leave. The UNHCR says 338,056 registered Afghan refugees have returned so far this year -- nearly six times higher than the number for the same period in 2015.
And just to remind you that the volume of undocumented migration globally is large, Joseph Chamie (former director of the UN Population Division) has a very informative article today summarizing what's going on around the world, to put the U.S. situation into context.
Excluding refugees who number more than 21 million and are under the protection of international conventions and agreements, it is estimated that of the remaining approximately 225 million migrants worldwide about 50 million are unauthorized migrants. The countries with the largest numbers of unauthorized migrants include the United States (11 million), India (at least 10 million), the Russian Federation (4 million), Malaysia (1 million) and the United Kingdom (1 million).
The U.S. and India are the 3rd and 2nd most populous nations in the world, respectively, and Russia is also in the top 10, so it is perhaps not too surprising that they also lead the league, so to speak, in terms of undocumented immigrants. Malaysia and the UK are not as populous, however, so the impact on those countries may seem disproportionate.

Monday, November 14, 2016

Political Demography at Play in the U.S. and Elsewhere

Underlying all of the demographic analyses about the U.S. election, as well as the Brexit vote, is that the demographic divide is about the "other." In the U.K., the (mis)perception that immigrants were undermining English society helped people (especially those with no real interaction with immigrants, as I noted at the time) to vote to leave the European Union. In the U.S., it was especially less educated whites who wound up giving Donald Trump the electoral college votes he needed from swing states in the midwest. Justin Stoler linked me to an NPR piece on what the exit polls taught us and this conclusion was particularly noteworthy:
... 6. And that leads to what might be the biggest story of the election — Democrats' cratering with blue-collar white voters.
Ohio and Iowa went by huge margins for Trump –- almost 10 points in Iowa and 9 in Ohio. Trump won Wisconsin and Pennsylvania (by less than a point), leads in Michigan (by an even smaller margin), and lost by less than 2 points in Minnesota.
These are all states that went for Democrats in six straight presidential elections. They were crucial to the Democratic Blue Wall, and Trump took a sledgehammer to it.
These are people who feel they are being forgotten by the country's "elite" and their anger has turned to immigrants and others who are different (classic xenophobia) and to those who support the "others." Donald Trump clearly fed into and hyped up that kind of rhetoric, but he didn't invent it. The demography of the U.S. and of Europe is changing and political leaders have paid insufficient attention to it. Why? Probably because the problems are large and complex. However, there is no going back on the demographic trends in place in the world--no matter how much some people might wish for that. Things are changing and politicians have to start coping with this reality--including the people who feel they are being left behind. This is what political demography is all about. My son, Greg Weeks, and I have contributed to the political demography literature in our book Irresistible Forces, which examines the forces underlying migration from Latin America to the U.S., and another good resource that everyone in politics should read is the book Political Demography: How Population Changes are Reshaping International Security and National Politics, edited by Jack Goldstone, Eric Kaufmann, and Monica Duffy Toft.

Saturday, November 12, 2016

The Uncertain Future of Women's Reproductive Rights in the U.S.

Among the many uncertainties associated with Donald Trump's election, especially when combined with the continued Republican majorities in both houses of Congress, is what will happen to women's reproductive rights in this country? Despite the fact that abortion has been legal in the U.S. for more than four decades, opponents have ramped up their attacks on abortion providers over time in a clear effort to scare away physicians and patients alike. One of the people sounding the alarm in a very personal way is Dr. Warren Hern, who runs an abortion clinic in Boulder, Colorado. Demographers know him for his excellent and ground-breaking work on fertility among the Shipibo people in Peru. That research helped him to earn a PhD in Epidemiology from the University of Carolina, Chapel Hill, but most of his professional life has been dedicated to helping women who needed an abortion. Yesterday, he published an online letter of concern:
I have survived, but some of my colleagues and friends have been murdered by anti-abortion fanatics.
Donald Trump has said that women should be punished for having abortions. Trump’s running mate, Indiana Governor Mike Pence, has used attacks on abortion as the principal focus of his public career.
In fact, one week ago I received a chilling letter from Representative Marsha Blackburn (R-Tenn.), the most ferociously anti-abortion member of Congress, who chairs the Select Investigative Panel on Infant Lives.
The letter demanded that I submit a wide variety of documents, including patient medical records in cases of gestations greater than 22 weeks. I have until Nov. 21 to comply.
The panel is looking for evidence that I am selling “baby body parts.” It has the power of subpoena and can cite me for contempt of Congress if I don’t comply by the deadline.
It is frightening. It is a witch hunt.
I am a physician helping patients, and I am being treated like a criminal.
The star chamber proceedings of the Select Investigative Panel on Infant Lives is a target identification program for the anti-abortion assassins. It is terrifying.
At a deeper level, I have no confidence that Trump and the current Republican leadership will protect my life and liberty because of the work I do to help women. Their public harassment of me increases my risk of assassination.
We should all be terrified that the situation is already this bad for physicians performing abortions. Whether or not you are in favor of abortion, the fact that it is legal means that no one should fear for their life to be either physician or patient--and it seems likely that the situation will get worse before (or if) it ever gets better. 

Friday, November 11, 2016

A House Divided: US Election Demographics

As the global shock of Donald Trump's election transitions to coping with it, data are emerging that remind us how markedly divided the country is geographically, demographically, and politically. The NYTimes put together exit poll data from nearly 30,000 people, including phone interviews with mail-in and early voters. These data obviously capture voters, so there is no issue with figuring out who is a likely voter. There may still be some non-response bias, but the in-person interview technique outside polling places almost certainly reduces that considerably. Anyway, here are three pieces of information that seem noteworthy to me: (1) a majority of men voted for Trump, while a minority of women did, but it was a pretty high minority of women; (2) the less educated you were, the more likely you were to vote for Trump, but Trump still grabbed almost half of white college graduates; and (3) as expected from the electoral maps and discussions ahead of time, the farther away you were from a city center, the more likely you were to vote for Trump.


The data also suggest that the win for Trump was as much about people not turning out to vote for Clinton, as it was about people voting for Trump. He got about the same number of popular votes as did Romney, but she got fewer votes than did Obama. Her team was worried that this might happen, and it did. Nonetheless, she still received more popular votes than did Trump, even though fewer electoral votes. This has happened five times in U.S. history--three times in the 19th century and twice in this century (Bush and now Trump).

Thursday, November 10, 2016

Is the Syrian War Changing the Country's Demographics?

Thanks to Abu Daoud for linking me to an interesting story from ABC News suggesting that President Assad of Syria is using the civil war there to somehow shift the country's demographics. Which demographic characteristic is at issue? That of loyalty to the government of Assad. I admit that I have never before thought of "loyalty" as a demographic characteristics, but there it is.
The opposition accuses the government of President Bashar Assad of using under-the-radar methods to discourage populations it sees as disloyal from returning, changing the demographics to help consolidate control over a corridor running from Damascus to the Mediterranean coast.
The government says it is doing all it can to bring people back.
"The main goal of the Syrian government is to return all displaced Syrians to their homes," National Reconciliation Minister Ali Haidar told The Associated Press last month.
More than 11 million people, nearly half Syria's population, have been driven from their homes by the war since 2011, including 5 million who fled abroad as refugees.
Without question, the entire world would breathe a collective sigh of relief for the civil war there to end in such a way that people could return and the country could be rebuilt. There is actually an implicit assumption that this will happen built into population projections by the UN Population Division and by the Population Reference Bureau, which assume that by 2030 the Syrian population will have increased from its current (maybe) 17 million to more than 25 million. 

Keep in mind that the birth rate in Syria had been dropping rapidly prior to the onset of violence, from 7 children per woman as recently as the 1980s to a level currently estimated to be slightly fewer than 3 children each. Of course, even at this lower level, children account for one out of every three Syrians, which is why photos of refugees always show so many children. However, a key demographic in the country is that fertility has been lower among the Alawites (an offshoot of Shia Islam and the religion of the ruling Assad family) and Christians than among the majority Sunni Muslim population. That is, in my opinion, the key demographic of the country.

Wednesday, November 9, 2016

The Power of Polls to Demographically Deceive Us...

Like many people I was very worried that Donald Trump might win the U.S. presidency, but I didn't really think it would happen. Why not? Because, like many people, I paid too much attention to the polls, even when I knew how flawed they can be. The problem is two-fold: (1) response rate to pollsters is very low (9% is a widely cited figure for the Pew Research polls), and (2) figuring out who is going to vote is difficult, even in the best of times. Dealing with both problems requires weighting schemes to adjust the collected data so that they presumably reflect the responses that you would have gotten with a very high response rate and a high level of voting certainty. The weighting is done almost entirely on the basis of demographic characteristics--age, sex, race/ethnicity, education, and place of residence being the major factors that are associated both with voter turnout and voter preferences. 

Where do the weights come from? Mainly from previous elections, based on demographic data from exit polls and from questions asked by the U.S. Census Bureau in the Current Population Survey.  If the weights are wrong, the results are wrong. Why might the weights be wrong? Because in the three most recent high profile cases of pollsters it getting wrong--Scottish independence vote, Brexit vote, and Donald Trump--there wasn't a sufficient history of voting in these unusual circumstances to allow pollsters to get the weights right.

I'm not blaming pollsters for Trump being elected. There are lots of reasons for that and experts will be debating those reasons for years to come. Pollsters just gave us incorrect information about what to expect and thus twisted our expectations about what was happening. We should have known better. Nate Silver's famous approach to the polling issue has been to average data from a lot of polls, focusing on those that seem--after the fact--to generally be closest to the truth about expected voting behavior. That seems to work reasonably well in more or less conventional situations, but at the moment we are living through unconventional times, and I expect that we are going to have more surprises ahead of us as the demographics of the country and the world continue to evolve.