You might have wondered why I didn’t pair my frequentist analysis in this post with a Bayesian one. Two reasons: length, and quite honestly I needed some time to chew over hypotheses. The default frequentist ones are entirely inadequate, for starters:
In chart form, their relative likelihoods look like this. [Read more…]
I’ve seen a few editorials along the lines of this one or this one:
An expanding working class may have dealt the final blow to the Kathleen Wynne Liberals, whose support has now become concentrated among upper class Ontarians and a shrinking middle class. For now, Doug Ford’s PCs hold a strong hand with working class voters—nearly double the support of either the Liberals or NDP. If Ford can sustain his working class support until June 7, he will sweep to a majority government.
It’s eerily similar to what we were hearing about Trump, isn’t it? And in Trump’s case, we’ve got a growing body of evidence that it wasn’t about economics, it was really about identity threat.
Donald Trump’s success in the 2016 campaign for the U.S. presidential election has defied the expectations of many Americans. This study is the first to demonstrate experimentally that the changing racial demographics of America are directly contributing to Trump’s success among Whites by increasing perceived threats to their group’s status. It is also the first to show that White Americans’ responses to increasing racial diversity depend on how identified they are with their ethnic group.
Major, Brenda, Alison Blodorn, and Gregory Major Blascovich. “The threat of increasing diversity: Why many White Americans support Trump in the 2016 presidential election.” Group Processes & Intergroup Relations (2016): 1368430216677304.
The most obvious finding in Table 1 is that, contrary to conventional wisdom, there is little to no evidence that those whose incomes declined or whose incomes increased to a lesser extent than others’ incomes were more likely to support Trump. Even change in subjective assessment of one’s own personal financial situation had no discernible impact on evaluations of Trump or on change in vote choice. Likewise, those who lost a job between 2012 and 2016 were no more likely to support Trump. […]
Mass opinion changes on status threat-related issues were not, by themselves, the driving force in increasing affinity for the Republican candidate. Instead, increasing relative distance from the Democratic candidate on threat-related issues, such as immigration and China, consistently predicted Trump support in a positive direction, whereas decreasing relative distance from the Republican candidate on trade and China also predicted change in the direction of voting for Trump. These consistently significant coefficients indicate that change over time in the candi dates’ perceived positions relative to those of individual respondents had a significant impact in increasing support for Trump. The pattern in Table 1 makes it clear that it was change in how the candidates positioned themselves on status threat-related issues combined with smaller changes in public issue opinions that predicted increasing support for the Republican candidate in 2016.
Mutz, Diana C. “Status Threat, Not Economic Hardship, Explains the 2016 Presidential Vote.” Proceedings of the National Academy of Sciences, April 19, 2018, 201718155. https://doi.org/10.1073/pnas.1718155115.
Ironically, a lot of the people banging on about the evils of “identity politics” are engaging in it themselves; people who identify with a privileged group are paranoid about losing those privileges, and will lash out at anyone who’s arguing for a more level playing field. But they recognize that “white people are inherently superior” and similar assertions are monstrous, so they need to hide behind some sort of coded language or flimsy argument. They aren’t part of an identity, oh no, so they can’t be engaging in identity politics! It’s not about a loss of privilege, oh no, it’s about working class voters feeling disenfranchised! It’s not about racism, oh no, it’s about immigrants taking our jobs!
You can see how this could explain the Ontario election: Ford was a horrifically unfit candidate, but identity-threat dog whistles could make him highly valued by a white majority that felt they were under threat, valued enough to transcend his flaws. The missing piece is whether he actually blew those whistles.
Earlier this month, the leader of the Ontario PCs generated significant controversy when he brought up the issue of immigration on the campaign trail, suggesting Ontario should “take care of our own” before worrying about immigrants. On a Google Hangout recently streamed by figures on the far-right linked to several Canadian hate groups, white nationalists appeared thrilled at the prospect of Doug Ford becoming Premier of Ontario. One avowed white nationalist even bragged that Ford was directly communicating with white nationalists using a “dog whistle” – ‘dog whistling’ refers to the use of coded language to disguise racist ideas to a general audience while signalling to racists themselves.
His recent attacks on Toronto “elites” and the media seem like conscious evocations of the bully-boy rhetoric Trump used to great advantage in his 2016 campaign. And his crude style – he once pledged to give his own party an “enema” – hasn’t backfired on him, yet. […]
In recent days, for example, he went out of his way to shun a planned leaders’ debate hosted by a Toronto-area black organization. Over the weekend, he drew boos during a speech to a Somali group during which he promised to reinstate an anti-gang police task force that had been intensely criticized for over-policing poor neighborhoods with large black populations (the Liberal government had pulled the plug on the task force).
Roger Keil, a York University urban geographer who studies politics and planning in Toronto’s suburbs, adds that when Ford was running for mayor, he didn’t exactly race to defend an opponent, a longtime local and federal politician named Olivia Chow, who came in for nasty racial attacks – just one of several that marred that election.“He never made an attempt to put himself between the racist and sexist overtones,” says Keil. “The real Doug Ford is a mean, clever strategist who has demonstrated time and again that he knows exactly what he’s doing.”
All signs point to yes. The best counter-argument isn’t to deny he made those whistles, it’s to argue Canadians are too savvy to fall for them. Click through on that last link and you’ll see what I mean; it goes to great lengths to point out that Toronto is very diverse, and that other conservative groups who used identity-threat dog whistles have been punished for it in the polls. All of them happened before Donald Trump, though. Trump may have shoved the Overton window so far that we’ve grown tolerant of certain whistles. Ford was also one of the first Canadian politicians to have a major press organization blast out and legitimize his whistles, much as Fox News did for Trump.
We need more data up here to be sure, but I suspect identity-threat dog whistles were enough to push Doug Ford over the top. Either way, they’ve become a major player in the North American political landscape.
Remember that post about the Ontario election? That trend in the polls continued, putting the NDP and PC’s in a dead heat. Meanwhile, Doug Ford was sued by his late brother’s wife, was recorded trying to sell fake party memberships to guarantee he’d become the PC leader, broke election law while fundraising, and on and on. Between all the lying and even promises of a blind trust to hold off allegations of nepotism, he’s about as close as Canada gets to a Donald Trump, which should have made an NDP minority a slam-dunk.
Led by Doug Ford, Ontario’s Progressive Conservatives have secured a majority government, ending nearly 15 years of Liberal power in the province. The NDP will form the province’s Official Opposition, while the embattled Liberals were handed a substantial rebuke from voters, losing the vast majority of their seats at Queen’s Park.
“Dead heat” in the polls, alas, didn’t take into account our first-past-the-post system; the vote distribution ensured the PCs held a seat advantage, though unlike the US this wasn’t due to gerrymandering. And there was a last-minute surge in support for the PC’s, so what was originally a 36-37% share of the vote grew to 40%. Ontarians just didn’t see the problem with electing a Trump clone.
On the plus side, Ford doesn’t have access to nuclear weapons and a military.
Yes, I’m reaching. *sigh*
Gawd, Enlightenment Now is turning into a slog. Let’s take one passage as an example:
Most surprises in history are unpleasant surprises, but this news came as a pleasant shock even to the optimists. In 2000 the United Nations laid out eight Millennium Development Goals, their starting lines backdated to 1990. At the time, cynical observers of that underperforming organization dismissed the targets as aspirational boilerplate. Cut the global poverty rate in half, lifting a billion people out of poverty, in twenty-five years? Yeah, yeah. But the world reached the goal five years ahead of schedule. Development experts are still rubbing their eyes. [Angus] Deaton writes, “This is perhaps the most important fact about wellbeing in the world since World War II.” [pg. 94]
That smelled like something worth researching, so I dug in. After plotting a counter-argument involving how that progress has been made, I read ahead to make sure I wasn’t taking Pinker out of context or missing something, and guess what?
Hundreds of millions of people remain in extreme poverty, and getting to zero will require a greater effort than just extrapolating along a ruler. Though the numbers are dwindling in countries like India and Indonesia, they are increasing in the poorest of the poor countries, like Congo, Haiti, and Sudan, and the last pockets of poverty will be the hardest to eliminate. Also, as we approach the goal we should move the goalposts, since not-so-extreme poverty is still poverty. In introducing the concept of progress I warned against confusing hard-won headway with a process that magically takes place by itself. The point of calling attention to progress is not self-congratulation but identifying the causes so we can do more of what works. [pg. 94]
Pinker admits that progress towards the Millennium goal of reducing poverty has been uneven, stepping on the argument I was going to make! Here’s what the UN’s official report has to say about that, in fact:
The world’s most populous countries, China and India, played a central role in the global reduction of poverty. As a result of progress in China, the extreme poverty rate in Eastern Asia has dropped from 61 per cent in 1990 to only 4 per cent in 2015. Southern Asia’s progress is almost as impressive—a decline from 52 per cent to 17 per cent for the same period—and its rate of reduction has accelerated since 2008.In contrast, sub-Saharan Africa’s poverty rate did not fall below its 1990 level until after 2002. Even though the decline of poverty has accelerated in the past decade, the region continues to lag behind. More than 40 per cent of the population in sub-Saharan Africa still lives in extreme poverty in 2015. In Western Asia, the extreme poverty rate is expected to increase between 2011 and 2015.
I keep tripping across similar situations. I’ll see Pinker make a point, run off to do research on it, then find he’s undercut his own point later. Which one sticks in your mind depends on your biases: if you think that we’ve made remarkable progress and can sit on the status quo, you’ll remember the part about cutting poverty and downplay the recognition that things have been uneven; if you think we’ve done well but can improve further, you’ll be thankful for the recognition that things could improve and might be tempted to forgive his fanciful flights.
But this can lull you from looking at the whole picture. That $1.25 a day was based on the poverty rates of the poorest 10-20 countries, yet the median poverty line in developing countries was $2 a day at the time. Standards of living are different in different countries, so while the UN may declare $13 a day “middle class” in the developed world, $4,748 a year doesn’t come close to the $7,320 you’d need for one bedroom in shared accommodations in Calgary. It’s nonsensical to use a single metric across the entire globe, and to be fair the UN probably realized this; the real goal was less about hitting specific targets and more of an effort to change hearts and minds, as well as open up wallets that were beginning to shut.
Nor were the results all that surprising. Pinker never considers why a set of goals agreed to in September 2000 could be so easily back-dated to 1990. In reality, the Millennium Development Goals were based on ideas that have been floating since the 1990’s, and their basic framework existed as early as 1996. The MDGs were developed by a broad coalition of organizations, and voluntarily agreed to by 189 countries. You wouldn’t have gotten that level of consensus had you proposed bold, risky targets to hit; the MDGs were designed to be achievable, from the very start. They were a PR stunt, albeit one that also saved lives and reduced misery.
Pinker almost leaves China out of the picture, but manages to sneak in this passage:
Cynical explanations, such as that the enrichment is a one-time dividend of a surge in the price of oil and other commodities, or that the statistics are inflated by the rise of populous China, have been examined and dismissed. [pg. 94]
Except, as I quoted earlier, the UN themselves admit that China’s progress was instrumental in reaching the Millennium goal for poverty. Pinker’s citation which tackles this head-on sets up a straw-person – “The global decline of extreme poverty – was it only China?” – then easily knocks that down by showing that China merely played a significant role in reducing the global poverty numbers.
If we’re to look at the two biggest successes for clues in reducing poverty, as Pinker suggests we do, we should then focus on becoming more autocratic and more liberal, more protectionist and more free-trade, plus more censorious and more censorious. Reality is much more complicated than he presents, and even the shades of gray he injects are less about honestly reflecting the real world than an attempt to distract and frustrate his critics. His grandiose pronouncements and simplistic graphs are proof of the latter. I mean, just look at this:
To generate that graph, you need to estimate the annual output of all human activity across two thousand years, then you need to correct it for differences in purchasing power and inflation. I’m surprised Pinker didn’t quote the guy who spanned a million years with similar techniques instead, he seems less cautious than the source Pinker used.
Limitations to data quality also means that estimating the growth of GDP per capita over many decades, or even centuries, is a hazardous undertaking that, despite the best effort of statisticians and researchers, will always be surrounded by a degree of uncertainty. As a result, earlier estimates of relative income levels diverge substantially from standalone benchmark comparisons or independent estimates of relative income for those early periods.
Income for pre-1500 populations were estimated by figuring out how many people were in extreme poverty, then assigning them an income of $1.25 a day. As Pinker himself notes, purchasing power adjustments are done by assuming things like loaves of bread and beer have a fixed value over all times and places. The global GDP was calculated by estimating the growth of GDP per capita per year, converting that into GDP per capita, then multiplying by world population estimates, a method the Maddison Project acknowledges can lead to “implausible results.”
The graph has so many assumptions built in it’s not worth the photons. Yet Pinker puts it in the front of his chapter on wealth, calling the chart “simple but stunning” and spending three paragraphs hyping it up. He then turns around and spends three paragraphs pointing out some (but not all) of its limitations, and concludes “A person whose wallet contains the cash equivalent of a hundred 2011 international dollars today is fantastically richer than her ancestor with the equivalent wallet’s worth two hundred years ago.” The 1818 equivalent of $100 is substantially less than the 2018 equivalent of that same $100?
An honest inquiry into poverty wouldn’t have featured that graph. But attempting to show that demands a helluva lot of research and dead ends, in Pinker’s case. Ugh.
South Korea has been a powerhouse in eSports. They’ve been running StarCraft tournaments since 2002, handed out most of the prize money, and as a consequence dominate the player rankings.
While distracting myself from a blog post, a video of a February 2018 StarCraft II caught my eye: a top South Korea player was playing… a Canadian woman? Huh? This sounded historic, so I tuned in to parts one and two, intending to keep it in the background while tapping away.
It pretty soon became the foreground. No spoilers for how the tournament turned out (the announcers totally sell it, anyway), but it was indeed a historic set of games. Non-Korean players rarely make it to the finals of a world championship, this tourney was a preview for eSports joining the Winter Olympics, and… I may not be much of a StarCraft player, but it looked like a damn fine round. Once you’re ready to be spoiled, no less than Rolling Stone explains several other ways those games were amazing.