April 2020

April 30, 2020 6:25PM

Decarceration in the Face of a Pandemic

America's jails and prisons are now among the deadliest environments on the planet. Most of them are desperately overcrowded, understaffed, unhygienic, and utterly unable to provide even minimally adequate medical care to those who contract COVID-19, which is now spreading like wildfire through those facilities, endangering not only the lives of prisoners, but also of guards, staff, and the communities to which they all return at the end of their shifts.

Thus, one of the most urgent—and contentious—debates in criminal justice today is over which prisoners to release in the face of a pandemic that is literally unprecedented during America's era of mass incarceration, which dates back to the early 1990s. Defense attorneys across the nation have filed a blizzard of early-release motions on behalf of their incarcerated clients, and the ACLU and other civil rights groups have sued a number of prisons and jails seeking the immediate release of particularly vulnerable inmates. Tragically, all of this is unfolding against the backdrop of a system that falls disgracefully short of meeting prisoners' medical needs during the best of times. In the midst of a genuine emergency, it is no secret what will happen to most people who contract COVID-19 behind bars: They will be left to live or die with only token medical attention.

As a result, all but the most obtuse proponents of mass incarceration now recognize that it has become morally indefensible to continue holding at least some fraction of the roughy 2.3 million people currently behind bars in an environment where we can neither adequately protect them from nor treat them for COVID-19.

But the system is having an extraordinarily difficult time deciding whom to release, and I think there are three key reasons for that: (1) we have become so cavalier in our use of the criminal sanction that the mere fact of a person's incarceration tells us nothing about his moral culpability or what risk his immediate release might pose to society; (2) we've become so inured to how horrible the conditions in jails and prisons are that exposing inmates to a new and exceedingly virulent pathogen may strike some as simply a marginal change in the already dismal circumstances of their confinement; and (3) thinking seriously about whom to set free and whom to keep behind bars in the midst of a pandemic raises questions that the carceral-industrial complex can scarcely afford to have people asking after the crisis subsides. I will address those points in turn.

Read the rest of this post »
April 30, 2020 4:14PM

Are Shutdown Orders a Taking? Part Two

States and cities have closed innumerable firms and ordered their workers and suppliers into idleness in response to the dangers of COVID-19 transmission. A month ago I posted about some legal questions raised as a result:

Are these takings of property for public use? If so, would the Supreme Court rule that they require just compensation under the Fifth Amendment’s Taking Clause? If not, is there nonetheless a case for some such compensation, such as emergency rescue payments, as rough justice?

I quoted two liberty‐​minded law professors on the subject. Cato adjunct scholar Ilya Somin wrote that on current Supreme Court precedent, which generally refuses to acknowledge most government regulatory action as a taking for Fifth Amendment purposes, courts would probably not award compensation for most shutdown losses. But both Somin and Keith Whittington went on to argue that the state of the law aside, there may be a moral case for compensating those who have been asked to forgo their livelihoods in response to a public safety emergency.

Since then there have been a couple of updates. Earlier this month the Pennsylvania Supreme Court ruled against several businesses that had argued (among several other theories) that being ordered to close was a compensable taking under the U.S. and Pennsylvania constitutions. Three dissenting justices held that the claims deserved more procedural consideration but did not endorse the application of a takings theory. Prof. Somin has more, noting that this outcome is to be expected under the current state of precedent. In recent days, Somin and business law professor F. E. Guerra‐​Pujol have had an exchange of views on whether the state of current law is really as discouraging for claimants as Somin believes. (For what it’s worth, I’d side with Somin.)

By coincidence, thanks to a casual reference in a piece by the Poynter news organization, I can report that interest in what you might call the moral argument for shutdown compensation go back at least a century. In 1906 the legislature of Massachusetts enacted a law to assure some support for laborers kept from their employ by public health orders in time of epidemic:

Section 95. If a disease dangerous to the public health breaks out in a town, or if a person is infected or lately has been infected therewith, the board of health shall immediately provide such hospital or place of reception and such nurses and other assistance and necessaries as is judged best for his accommodation and for the safety of the inhabitants, and the same shall be subject to the regulations of the board. The board may cause any sick or infected person to be removed to such hospital or place, if it can be done without danger to his health; otherwise the house or place in which he remains shall be considered as a hospital, and all persons residing in or in any way connected therewith shall be subject to the regulations of the board, and, if necessary, persons in the neighborhood may be removed. When the board of health of a town shall deem it necessary, in the interest of the public health, to require a resident wage earner to remain within such house or place or otherwise to interfere with the following of his employment, he shall receive from such town during the period of his restraint compensation to the extent of three fourths of his regular wages; provided, that the amount so received shall not exceed two dollars for each working day. [emphasis added]

By the standards of that time, two dollars a day is not as ungenerous a substitute wage as it may seem from our remove; if this guide to 1904–1905 wages is accurate, it would have sufficed to replace three‐​quarters or more of wages in all but the most highly skilled of Massachusetts industrial trades.

Nothing in the policy world is ever as new as it may seem, including measures to tide over the workforce whose livelihood has been interrupted by an epidemic.

April 30, 2020 4:10PM

A Little Advice for Efficient, Effective, Online Higher Ed

I’ve worked in higher education since 1987 when I started as an Assistant Professor of Management at Oklahoma State University, with a focus on strategy, innovation, and entrepreneurship. About 15 years ago I narrowed my research to one industry—education. Then, about eight years ago, I became intrigued by the potential of online and blended education as a tool to drive improvement.

My experience with distance education started well before I became a professor. Fifty‐​two years ago I took my whole sophomore year of high school from the University of Nebraska while living in Oshogbo, Nigeria. Educationally it worked great. The only problem was the communication time delay caused by the Nigerian mail system. It took longer for mail to travel the 121 miles between Lagos and Oshogbo than the 6,448 miles between Lagos and Lincoln.

As a result of my positive student experience, I’ve regularly volunteered to teach distance courses using the communication method of the time—from mail correspondence, to compressed video, to online. Over the last several years, half my teaching was done online in a recorded video/​discussion board/​major written project format, and the other half blended.

Over time I became convinced college costs could be dramatically cut, and quality improved, by moving to online and blended delivery. I discussed this a bit in my 2011 Cato Policy Analysis “Federal Higher Education Policy and the Profitable Nonprofits.”

Three years ago I took emeritus status at Oklahoma State and started TEL‐​Education, a non‐​profit educational publishing and technology organization. TEL’s goal is to make affordable, high‐​quality education available to anyone, anywhere, at any time. We develop courses for colleges and high schools to use in providing general education and early college high courses.

In our three years we’ve gone through several iterations of business and instructional models, publishing processes, and technology solutions, and have worked with multiple colleges and high schools. Here are some things we’ve learned from our experiences so far, and from other providers of distance education, that could be of help to more traditional colleges and universities as they move into a COVID-19 era that may force them to deliver more education long distance:

Online courses must be highly engineered in order to scale. As with traditional face‐​to‐​face courses, online courses should be designed to maximize learning. At TEL we design individual lessons to provide a clear context for each course concept, and then move students through the learning stages of information elaboration, relevance, agency, and mastery. This design allows students to absorb the information more readily and apply it in meaningful ways. Good course design is vital no matter the delivery method. As many students may be experiencing this semester, a bad face‐​to‐​face class doesn’t’ suddenly become good when moved online.

Unfortunately, you can’t simply move a good in‐​person class online, particularly if you are wanting to scale. You also need an online learning system that reinforces learning and has processes and technology that are easy for students to navigate. This requires a technology platform, like EdX or Coursera, that is built specifically for the delivery of large‐​scale online courses.

Self‐​paced, online courses can be delivered at an extremely low cost. For example, Straigtherline offers an a la carte, on‐​demand program that costs $50 per credit hour, while our Courses on Demand are similarly priced. These courses allow students to earn college credit that can be used towards degrees at most American colleges and universities, including big public research universities and most private colleges.

Colleges have begun offering their own self‐​paced courses at discounted prices. York College is a leader in this regard, allowing students anywhere the ability to earn an A.A. degree while in high school for $67 a credit hour.

Online early college high school (concurrent or dual enrollment) works well in any setting. Early college high school isn’t just for high schools located close to a college or university. Self‐​paced online programs can serve high school students anywhere. The traditional public high school in Pawhuska, OK, offers such a program to its students. Pawhuska is a small, rural school (50 graduates a year) with a diverse student body. (It’s the Pawhuska of the Killers of the Flower Moon book and upcoming movie).

They currently offer 16 different courses to their students in both semesters.They can do this because an accredited college, not the high school, provides the instruction. What the high school provides is the support system most students, particularly the economically disadvantaged, need. Their role is providing a mix of encouragement and accountability to every student.

There is a cure to the college cost disease. Self‐​paced online means that college costs can be radically reduced. While residential college will remain more expensive than online, its cost can be also be radically reduced by moving to blended delivery over pure face‐​to‐​face, and mixing in some self‐​paced online courses.

Most dramatic are the savings from online early college high school. Online early college high courses can be delivered for less than the cost of regular high school courses. With online early college high, an associate’s degree achieved in high school is possible for any good student. That enables them to shave two years off their educational journey, a dramatic savings in both tuition and opportunity cost.

The Post‐​Crisis Future. While relatively short‐​term in nature, the current COVID disruption in education highlights huge future trends, including: 1) The first part of college will be pushed down into high school, so students will be buying fewer years of “college experience”; and 2) students will become more demanding buyers of education, both in terms of quality and cost. Colleges must adjust their services and pricing accordingly.

April 30, 2020 2:58PM

COVID-19 Should Make Us Grateful for Technology

In this National Review essay, I take a look at the impact of technology on health care, food supply, work, and sociality in the time of COVID-19. Here is a part of it:

Humanity lived with deadly diseases for millennia without fully knowing what they were, how they were transmitted, and how they could be cured. The fate of humanity, our ancestors thought, fluctuated under the extraneous influence of the “wheel of fortune” and there was nothing that anyone could do about it. One day you were alive and next day you were not. Contrast that glacial pace of progress, and the fatalistic acceptance of disease and death, with our response time to the current pandemic. The Wuhan Municipal Health Commission reported the existence of a cluster of cases of “pneumonia” in Wuhan on December 31. On January 7 the Chinese identified the pathogen (novel coronavirus) responsible for the outbreak. On January 11 China sequenced the genetic code of the virus, and the next day it was publicly available. That enabled the rest of the world to start making diagnostic kits to identify the disease. To take one example, the first COVID-19 infection in South Korea was identified on January 20. On February 4, the first test kit (made by Kogene Biotech) entered production. On February 7, the test kit was available at 50 locations around the country. Other countries followed suit. The World Health Organization, which declared COVID-19 a global pandemic on March 11, may have acted too late. Still, it is noteworthy that just two months expired between the first sign of trouble and the time when the entire world put measures in place to retard the spread of the disease.

Please don’t forget to check out this short Human Progress video on the treatment of contagious diseases throughout history.

April 30, 2020 2:57PM

Epidemiological Models Can’t Tell States When to Open or How

With more than a dozen states starting to relax mandatory social distancing edicts, some organizations are offering advice to states on when they might prudently alleviate or decentralize current mandates. The most ambitious effort is an offshoot of a model from the Institute for Health Metrics and Evaluation (IHME) at the University of Washington. Having been made famous in White House briefings by Doctors Birx and Fauci, that “Murray Model” could conceivably be influencing federal and state official about the proper timing and design of state decisions to relax mandates.

I assembled a table “Evaluating State COVID-19 Rules” to examine IHME recommended dates for states to begin “opening up” businesses, shopping, recreation and employment, while adding two sources of relevant facts.

Evaluating State COVID-19 Rules

Deaths

per Million

Reproduction

Rate (Rt)

IHME

Opening Date

IHME

Score

AK

12

0.82

N/A

5

AL

50

0.85

19‐​May

5

AR

19

0.99

28‐​Jun

3

AZ

42

1.01

6‐​Jul

3

CA

48

0.9

20‐​May

5

CO

133

0.9

1‐​Jun

5

CT

583

0.88

17‐​Jun

4

DC

278

0.93

27‐​May

5

DE

144

1.08

20‐​May

5

FL

57

0.79

21‐​Jun

4

GA

101

0.81

28‐​Jun

4

HI

11

0.79

11‐​May

5

IA

43

1.06

1‐​Jul

4

ID

36

0.84

19‐​May

5

IL

166

0.91

21‐​May

5

IN

149

1.05

22‐​May

5

KS

44

1.05

28‐​Jun

3

KY

51

0.9

22‐​Jun

4

LA

386

0.76

26‐​May

5

MA

462

0.9

21‐​Jun

4

MD

169

0.92

27‐​May

5

ME

38

0.97

19‐​May

5

MI

358

0.69

21‐​May

5

MN

54

1.06

7‐​Jun

4

MO

54

0.89

17‐​Jun

4

MS

80

0.99

1‐​Jun

5

MT

14

0.86

18‐​May

5

NC

36

0.95

13‐​May

5

ND

19

1.09

20‐​Jul

2

NE

29

1.14

7‐​Jul

3

NH

45

0.97

17‐​May

5

NJ

725

0.91

28‐​May

5

NM

53

0.94

5‐​Jun

4

NV

77

0.94

23‐​May

5

NY

1180

0.83

28‐​May

5

OH

69

0.92

15‐​May

5

OK

53

0.85

N/A

4

OR

24

0.91

30‐​May

4

PA

161

0.93

28‐​May

5

RI

226

0.93

22‐​Jun

4

SC

39

0.94

14‐​Jun

4

SD

13

0.89

2‐​Jul

2

TN

28

0.9

1‐​Apr

4

TX

25

0.74

15‐​Jun

4

UT

15

0.9

7‐​Jul

3

VA

58

0.93

27‐​May

5

VT

75

0.8

24‐​Mar

5

WA

108

0.88

31‐​May

5

WI

52

0.99

22‐​May

5

WV

21

0.81

10‐​May

5

WY

12

1.08

1‐​Jun

3

The first column in the table shows COVID-19 deaths per million by state. The national average death rate is 179 per million, yet 31 of the 50 states have experienced fewer than 60 deaths per million. On the basis of experience, these 31 states (at least less‐​populated counties within the states) appear to be relatively low‐​risk places to experiment with relaxing restrictions.

The second column shows the Reproduction Rate (Rt) in the latest week. In this series, “(Rt) represents the effective reproduction rate of the virus.” That is, if each person affects one other then Rt equals 1. The curve is then flat, since everyone cured is matched by someone newly infected. While Rt is below 1.0, on the other hand, that means new cases are smaller than the number cured so the number still infected is declining.

By the week ending April 25, 42 states had an Rt below 1.0. Among some of the first to lighten restrictions, Rt was down to 0.83 in Florida and Oklahoma by April 30, 0.78 in Georgia and 0.75 in Texas.

Soon after famously projecting that U.S. deaths would total only 60,000­ –a number already exceeded­– the IHME model’s latest update “raises the outlook for the cumulative U.S. death toll through Aug. 4 … to a new figure of 74,073.” The new curve projects that nationwide daily deaths will drop to 0–10 by July 1. But that is said to depend on maintaining “current social distancing… until infections [are] minimized and containment implemented.” That may explain why these modelers recently began advising states what to do and when to do it. “A recent analysis from the Institute for Health Metrics and Evaluation,” noted an April 22 local news story, “suggests that Oklahoma may be among several states that need to continue social distancing until late June or early July.” Oklahoma Governor Kevin Stitt ignored that advice and allowed haircuts.

The IHME team created a 1–6-point scorecard awarding the most points for the most severe “current social distancing” rules. They use these scores to propose specific dates in May‐​July when each state is granted approval to begin graduating from business and school closures and universal home isolation to more‐​targeted “containment strategies… [such as] testing, contact tracing, isolation, and limiting gathering size.”

The table’s third column “IHME Opening Date”– shows when “relaxing social distancing may be possible with containment strategies.” The last column shows each state’s “IHME Score,” which I argue is mainly what determines each state’s date for relaxing “social distancing” as they define it.

The modelers “evaluate [State] government‐​mandated social distancing measures… based on the New Zealand Government alert system Level 4″ (which was eased to Level 3 on April 28). They rate the virtue of states by how many of six equal‐​weighted broad categories of restrictions they implemented. A perfect score of 6 would require (1) closing nonessential businesses, (2) closing nonessential services, (3) closing all schools, (4) banning large gatherings, (5) issuing stay‐​at‐​home orders, and (6) “travel severely limited.”

No state scored more than 5, because no state could meet the New Zealand standard for domestic or international immobility. New Zealand banned foreign travelers on March 12 (because 5 people died out of 4.8 million) following other island nations Hong Kong and Singapore. New Zealand’s March 23 Level 4 plan also decreed “Public transit is closed.” Period.

Since no U.S. state lived up to New Zealand’s travel and transit bans, that left the highest possible score at 5. As a result, the unparalleled risks of proximity to strangers in New York’s subways, trains, busses and airports became irrelevant to scorekeepers who granted New York the same 5‐​point safety score as states with no mass transit or international airports. So, the IHME gives New York its blessing to relax social distancing after May 27. All others states that checked five boxes are likewise sanctioned to open in May, some a week or two earlier.

States with a score of 4, on the other hand, have to wait until June before doing anything different. And states with a score of 2 or 3 have to wait until July when the IHME model predicts zero COVID-19 deaths nationwide (we may not believe that, but they should because it’s their model).

In short, the timing recommended for opening a state’s economy depends on points earned on the basis of adherence to New Zealand’s formerly tough rules. However, the IHME makes a contrary claim that the “timeline is based on the date by which our model projects that COVID-19 infections will drop below 1 per 1 million people.” Yet the model projects only deaths per day, not “infections” (most of which are asymptomatic and uncounted) or even cases (which are sensitive to testing). Also, the model tracks actual deaths, not deaths per million.

Even if the dates for relaxing social distancing depended on projected deaths falling to 1 per day (rather than infections per million) that goal has been met by South Dakota ever since April 8 which you might think would make the state safe to open up entirely. South Dakota never closed anything but schools and large gatherings, in fact, which is why it is cursed with the lowest IHME score of 2. So, this remarkably safe state is admonished by the IHME to keep its two little rules in place until “after July 5, 2020.”

The fact that some states have a larger number of social distancing rules cannot explain why some had fewer deaths than others. On the contrary, South Dakota and 18 other states had fewer than 60 deaths per million regardless of their sub‐​par IHME scores of 2–4 and their resulting late dates for changing current rules in any unapproved way (such as confining home isolation to the sick and vulnerable)

Policies to deal with epidemics should be evaluated more on actual results and less on illusions that anyone really knows which schemes will prove more cost‐​effective and durable than others.

After graphically comparing mitigation policies with results among many states and countries, and finding no clear connections, Yinon Weiss concluded that “legal shut downs may only play a small role in the slowing down of the spread when compared to other factors such limiting travel to/​from hot spots, degree of public transportation use, washing hands, cancelling high transmission events such as conferences, and whether people are staying home when sick or exposed.”

The first columns of my table provide two relevant measures to discern how well states are actually coping with COVID-19, rather than relying on malleable model projections or subjective points scored in the IHME 1–5 virtue award contest. New tests, cases, hospitalizations and deaths also bear watching for potential hot spots, of course. Thankfully, no infectious disease has ever before been so quickly and transparently monitored by so many private and state online trackers.

If federal or state officials are instead considering relying on dates and scores from the University of Washington’s IHME team to decide what to do and when to do it, they should carefully reconsider delegating such authority to unaccountable technocrats to make social, health and economic choices that will deeply affect many millions of American lives.

April 30, 2020 12:17PM

E‐​Verify Errors Plague Workers with Temporary Protected Status

We’ve long complained about E-Verify’s failures. It fails miserably as a system designed to exclude illegal immigrant workers from employment. It is also error‐​prone, affecting some native‐​born Americans but mostly flagging legal immigrant workers as illegal immigrants. Correcting those errors can take weeks if all of the bureaucratic steps are followed. It’s no surprise, then, that E‐​Verify especially excludes workers on Temporary Protected Status (TPS) from working. Despite being authorized to work legally in the United States, TPS migrants are disproportionately flagged as ineligible to work by E‐​Verify. This is further evidence that E‐​Verify is a dangerously flawed system that should be repealed.

Background

E‐​Verify is an electronic government system that is supposed to stop employers from hiring illegal immigrant workers and thus reduce the main benefit of coming to the United States in the first place: high wages. E‐​Verify is supposed to work this way: Employers enroll in E‐​Verify and when they hire somebody, the employer is supposed to run that person’s identity information through E‐​Verify to confirm that he or she is legally eligible to work in the United States.

Most legal workers are instantly approved to work while some workers, mostly those who are illegal workers, are flagged with a tentative non‐​confirmation (TNC). The worker and the employer then have a certain amount of time in which to resolve the TNC by fixing errors entered into the E‐​Verify portal, providing other forms of identification, or pointing out errors in government databases that produce a mismatch. If they can resolve the TNCs, the worker can remain employed. If they can’t, E‐​Verify issues a final non‐​confirmation (FNC) and the worker must be fired.

How E‐​Verify Affects Workers on Temporary Protected Status (TPS)

E‐​Verify interacts with migrants on Temporary Protected Status (TPS) in a particularly pernicious way. TPS is a status for migrants in the United States who can’t return to their home countries due to an ongoing armed conflict, an environmental disaster, or other extraordinary and temporary problem. TPS migrants are not removable from the United States, are granted employment authorization documents, and may be granted travel authorization. There are approximately 411,000 migrants on TPS from 10 countries: El Salvador, Haiti, Honduras, Nepal, Nicaragua, Somalia, South Sudan, Sudan, Syria, and Yemen. TPS status will expire for many migrants in the coming years due to decisions by the Trump administration.

According to a recent USCIS data release in response to a FOIA, E‐​Verify was run 17,909 times against TPS migrants by employers in the 4th quarter of fiscal year 2019. Of those 17,909 E‐​Verify queries run against TPS migrants, E‐​Verify approved 16,299 of them to work and issued 1,610 with a TNC. In other words, about 9 percent of the E‐​Verify cases run against those on TPS in the 4th quarter of 2019 were mistakenly labeled at TNCs.

E‐​Verify later rescinded the TNCs in 323 (20 percent) of the E‐​Verify cases against TPS migrants, thus allowing them to work. About 9 percent of the legal TPS migrant workers were initially flagged by E‐​Verify as illegal and only 20 percent of those initially flagged were eventually allowed to legally work. About 71 percent were not resolved because the worker didn’t challenge the TNC or, more likely, the employer didn’t inform the worker of a problem. E‐​Verify incentivizes employers to pre‐​screen applicants by running them through E‐​Verify before hiring them, in contravention of the law, so many TPS workers hit with TNCs will never have a chance to resolve them.

During the 4th quarter of fiscal year 2019, there were 9,824,265 E‐​Verify cases with only 176,837 flagged as TNCs. That 1.8 percent TNC rate for all E‐​Verify queries is about one‐​fifth as high as the rate of E‐​Verify queries coming back as TNCs for TPS migrants run through E‐​Verify. Of all E‐​Verify queries in the 4th quarter of fiscal year 2019, 13.5 percent of erroneous TNCs were later approved for work. E‐​Verify queries run on TPS migrant workers resulted in an erroneous TNC rate 48 percent higher than the nationwide erroneous TNC rate. TPS migrants are about 0.13 percent of the U.S. population but they accounted for 0.9 percent of all TNCs and about 14 percent of all resolved TNCs – far in excess of their share of the population.

TPS migrants were about 7.3 times as likely to be hit with a TNC than we’d expect if E‐​Verify errors were distributed randomly. TPS migrants were about 10.8 times as likely to have their erroneous TNC corrected as the rest of the population run through E‐​Verify. If 100 percent of the TPS migrants had their erroneous TNCs corrected, they would account for 6.7 percent of all corrected TNCs that resulted in the worker being hired, which is a rate of correction that would be 54 times greater than what we’d expect if E-Verify’s errors were distributed randomly.

TPS migrants were more likely to be run through E‐​Verify than their share of the U.S. population would suggest. They account for about 0.13 percent of the population but 0.18 percent of all E‐​Verify queries in the 4th quarter of fiscal year 2019. Even taking that into account, TPS migrants are still more likely to be granted false TNCs. No matter how you slice the numbers, even correcting for their elevated share of E‐​Verify queries, TPS holders disproportionately bear the costs of the error‐​prone E‐​Verify system.

The results for the 4th quarter of 2019 are not an anomaly. During the entire time covered by the USCIS data from January 2016 through January 21, 2020, there were 191,990 E‐​Verify queries run against TPS migrants. Of those, 11,610 or 6 percent, came back as TNCs. Of the 11,610 TNCs, 17.1 percent were later approved for work. About 83 percent of TNCs against legal worker authorized TPS holders were not corrected during that time.

During the Obama administration, from January 2016 through December of that year, 1.95 percent of E‐​Verify queries against TPS workers resulted in a TNC and 15.8 percent of them were erroneous. During the Trump administration, from January 2017 through late January 2020, 7 percent of E‐​Verify queries against TPS workers resulted in a TNC and 17.1 percent of them were erroneous. Although the rate of confirmed erroneous TNCs was similar under both administrations, the initial TNC rate was more than three times higher under the Trump administration than under the last year of the Obama administration. Furthermore, the average monthly number of TNCs issued to TPS migrants increased from 59 during the last year of the Obama administration to 301 during the Trump administration.

E‐​Verify has been updated at least twice in ways that may have affected the issuance of TNCs for TPS workers. The first was in May 2018, right after a spike in the number of TNCs delivered against TPS migrants, and it guaranteed that E‐​Verify queries were run simultaneously against the Department of Homeland Security and the Social Security Administration databases. A second update in December 2019 automated the resolution of more TPS cases at the initial step, thus reducing the need to perform additional verification for cases involving TPS beneficiaries. It’s too soon to see if this later update has had any effect.

Conclusion

We’ve long known that E‐​Verify is an error‐​prone program that unintentionally denies legal immigrants the ability to work legally, but now we know that TPS workers are at even greater risk of being erroneously targeted by this program. At a very minimum, E‐​Verify errors like these should be minimized before the system is adopted by other states and localities. This is another reason to oppose E‐​Verify.

April 30, 2020 12:13PM

Darn Those Greedy Insurance Companies

Reuters reported the other day that the savings from forgone health care and claims due to people sheltering in place is enough to pay for the cost of full coverage for COVID-19 care that many insurers are offering:

“The costs from COVID-19 are going to be actually very small and more than outweighed by the deferral of elective procedures. The net impact is going to be positive for them,” said Jeff Jonas, portfolio manager with Gabelli Funds…

The largest U.S. for‐​profit hospital operator, HCA Healthcare Inc, said it has seen a 70% drop in outpatient surgeries so far in April compared with a year ago, while inpatient admissions declined 30%…

Health plans and employers who provide insurance have seen an overall decline in healthcare use of about 30% to 40% excluding COVID-19 patients, according to Tim Nimmer, the global chief actuary at Aon, a benefits company that advises large corporations and health plans.

“For each month that this goes on, we are expecting about 1.5% to 2% in annual costs to be reduced,” Nimmer said. “The number one issue is how long will this go on.”

As for what health insurers will do with the money they’re saving, the ACA’s “medical loss ratio” provisions require that if insurers (large‐​employer plans) spend less than 80 percent (85 percent) of premium revenue on claims, they have to refund that money to enrollees. Since insurers are spending far less than expected on claims, we can expect the ACA will require them to send lots of rebate checks to enrollees.

But do we need government regulation to ensure that insurers return such windfalls to their enrollees? Consider what’s happening in the auto insurance market.

Similar to health insurers, auto insurers are saving money because people are driving less. The Wall Street Journal reports, “Allstate Chief Executive Tom Wilson said mileage was down ‘an unprecedented’ 35 percent to 50 percent across the U.S. since mid‐​March, including in states without shelter‐​in‐​place restrictions.” (That latter part provides a helpful reminder that government affects pandemic response only at the margin.) Less driving means fewer accidents, so claims have likewise fallen as much as 40 percent.

What are greedy insurance companies doing with those windfall profits? They are issuing refunds.

Policyholder‐​owned American Family Mutual Insurance is refunding $200 million to policyholders. Publicly traded Allstate is refunding $600 million to customers. American Family COO Telisa Yancy explained, “It is real dollars we expected to pay out this year and no longer have to pay out. We are sharing it back right now when our customers probably most need it.” The WSJ reports, “Wall Street analysts believe many will follow suit…for reasons including competitive pressure.”

Darn those greedy insurance companies.