Numlock News: December 12, 2018
By Walt Hickey
Liking Numlock? Forward today’s email to a friend you think may enjoy it and might subscribe.
The global beauty market is worth $532 billion, and Glossier Inc. — the rare company that does almost the entirety of its business on the internet — is now poised to grab $100 million of that this year. The company began as a blog in 2010 and, $86 million in venture capital funding later, could be on the cusp of an international public offering. The beauty business is one of fierce competition and while rival retailer Ulta has seen upwards of 20 percent revenue growth in the past three years, anyone who’s making it happen in both the cutthroat worlds of cosmetics and Instagram marketing has to be doing something right.
The world spends about $12 billion on paper towels for home use annually, setting aside use in public bathrooms or offices. While many other countries use reusable washcloths or mops — you know, like chumps — the undisputed champs of single-use cleanup problems is unquestionably the United States. It’s not even close: the U.S. spent $5.7 billion on paper towels in 2017, with nearest runner-up France spending $0.635 billion. In 2017, the average American spent $17.50 on paper towels. As a millennial, I must ask what else are you supposed to use as a napkin?
Federal tax collectors have seen their budget to investigate tax cheats gutted and it’s having enormous effects on the government’s ability to collect revenues. While nobody wants to get audited, this is letting people who straight-up skip filing a return get off scot-free. Investigations of non-filers dropped from 2.4 million in 2011 to 362,000 last year. That reduction results in $3 billion in lost revenue every year. Collections from people who do file but don’t pay have also fallen, and those expire after 10 years if they’re not pursued. In 2010, $482 million in such tax debts lapsed. In 2017, that figure was $8.3 billion.
Researchers are abandoning academia earlier and earlier than ever. An analysis of scientific publishing careers — seeing when someone first appears in scientific literature and then when they stop — shows that people are rotating out of the publishing game earlier and earlier. In 1970, the half-life of a publishing career was about 35 years in astronomy and 30 years in ecology. In 1990, when robotics shows up on the scene, that had fallen to just under 25 years, just under 20 years, and in the case of robotics 15 years, respectively. Today more than half of researchers quit the field within five years in all cases. I have many theories as to why we’re seeing people leave astronomy, ecology and robotics so quickly — namely alien abductions, vicious predators with a taste for ecologists, and killbots — but, hey, maybe modeling the structure of the academic labor force after feudalism was a mistake you know?
A Nestlé plan to increase the amount of Michigan water it pumps from the ground from 250 gallons per second to 400 gallons per second was met with widespread opposition from Michiganders. The state’s Department of Environmental Quality said initially it received 75 comments in support of the move and 90,945 formal public comments opposed. A Freedom of Information Act request asking to see the 75 comments actually prompted a revision: 18 were moved to other categories upon review and 13 of the remaining 57 positive comments were, in fact, duplicates. Nestlé’s request was granted in April.
About 10 percent of the non-elderly population is uninsured in the United States. Many of those people are eligible for free health insurance from the ACA if they would only apply, but the current administration is not attempting the outreach that the Obama administration did. About 4.2 million uninsured people could get a bronze ACA plan in 2019 and, after the tax credits they’d be eligible for, would pay $0 in premiums. That means an estimated 27 percent of 15.9 million uninsured could shop on the marketplace and have a free bronze plan in 2019. So tell your friends, because the government has no plans to.
The census is attempting to make the block-level data it makes available to the public a little less accurate after an internal study realized that new data processing techniques may make it possible to identify specific people based off this data alone. This goes against the anonymity that underscores the mission of the bureau. For instance, data tables published after the 2010 census had nearly 8 billion numbers in all, or roughly 25 numbers for each resident, despite the fact only 10 questions are asked. The census has run hypothetical attacks on itself, essentially trying to figure out how much someone could reconstruct from the public data alone. They then compared those reconstructions to the actual, confidential private records. The result? Roughly 50 percent were an exact match and for 90 percent of people there was at most one mistake. So, now they’re going to get ahead of that in 2020 and fuss it up a bit before putting it all out there.
Thanks to the paid subscribers to Numlock News who make this possible. Subscribers guarantee this stays ad-free, and get a special Sunday edition. Consider becoming a full subscriber today.
Thank you so much for subscribing! If you're enjoying the newsletter, forward it to someone you think may enjoy it too!
Previous Sunday special editions: Signal Problems · CTE and Football · Facebook · Shark Repellent · Movies · Voting Rights · Goats · Invitation Only · Fat Bear Week · Weinersmith · Airplane Bathrooms · NIMBYs · Fall 2018 Sports Analytics · The Media · Omega-3 · Mattress Troubles · Conspiracy Theorists · Beaches · Bubbles · NYC Trash · Fish Wars · Women’s Jeans · Video Stores
Send links to me on Twitter at @WaltHickey or email me with numbers, tips, or feedback at firstname.lastname@example.org. Send corrections or typos to the copy desk at email@example.com.