Archives: 05/2018

Trump Officials: We Want to “Enforce the Laws” But Only Certain Ones

Testifying before Congress this week with the heads of the other two immigration agencies, Immigration and Customs Enforcement head Thomas Homan asserted that “no one on this panel is anti-immigrant,” claiming that while he “feels bad for some of these people,” they are just “enforcing the law that Congress enacted.”

Yet he and his copanelists spent much of the hearing arguing for not enforcing the law, but rather changing it to make it more anti-immigrant than it already is. His testimony repeatedly bemoaned “Congressional inaction” that requires him to enforce laws that allow, in his view, too many asylum seekers to apply for asylum in the United States.

Homan also doesn’t want to enforce the law that prohibits him from jailing women and children pending their asylum hearings and wants more money and authority to jail other immigrants as well. These are not the comments of a disinterested, passive “enforcer of the law,” but of a policy advocate who believes that current laws are too friendly to immigrants.

His copanelists were just as exuberant in their desire for nonenforcement. Customs and Border Protection chief Ronald Vitiello’s testimony told Congress that he doesn’t want to enforce the law that gives hearings to little children who arrive at the border by themselves from Central America. He said that he “remains committed to working with Congress to address these issues in support of the priorities of this Administration”—in other words, his priorities differ from current law.

No member of the administration was more hostile to enforcing current law than U.S. Citizenship and Immigration Services Director Francis Cissna. He was outraged that he cannot refer for criminal prosecution asylum seekers who have their applications denied. It’s not enough, in his view, for them to be denied safe haven and kicked out of the country. They should face jail time for filing what he believes are—but current law does not consider to be—“frivolous” applications.

Cissna came ready with a laundry list of complaints against current law. He commanded Congress to “immediately pass legislation” to fast-track deportations of women and children seeking asylum, to make it more difficult for people to be able to even apply for asylum in the first place, and then to revoke asylum to anyone who visits their home countries.

He complained that current law prohibits deporting “victims of gang violence,” which is just too “generous” in his view. “Someone only has to show that there is a ‘reasonable possibility’ of suffering persecution on account of a protected ground in order to qualify for asylum,” he lamented. All of these laws need to be changed, not enforced, he believes.

At one point, Cissna even endorsed legislation by Rep. Bob Goodlatte (R-VA) that would slash legal immigration to the United States by nearly 40 percent and kick out of line millions of legal immigrants who have been waiting, sometimes for decades, for the ability to come legally to the United States. He doesn’t want to enforce the law if it leads to more immigrants coming in.

This swath of policies are impossible to describe as anything other than anti-immigrant, and they have nothing to do with anything Congress has passed. They are a wish list of the most extreme version of the anti-immigrant agenda, targeting not only those in the United States without legal status, but also those who have it or wish to have it.

In a moment of eloquent incoherence, Cissna told the panel, “There is only so much we can do as a Department to enforce the rule of law when serious loopholes exist within current law.” In other words, we don’t want to enforce the law as it is right now, so change it. But he doesn’t want to change it in a way that makes it easier for people to follow the law, but rather makes it more difficult.

“Just enforce the law” is a mantra in favor of inaction, of making the status quo sacrosanct, but no one believes the status quo is perfect. Homan and his colleagues admitted as much. But they refuse to own up to the fact that their desire to not enforce current law is uniformly hostile to immigrants—legal or not. If that’s not “anti-immigrant,” what is?

Show Me the (Education) Money, Finale!

Long-Term, National: Money and Employees Have Poured In

Now that we’ve looked at scads of data—on spending, staffing, salaries—what can we conclude about the state of resources in public schools?

First, we need to recognize that the period since the Great Recession has been an anomaly in nearly a century of education spending. Whether in total or on a per-pupil basis, until the Great Recession we rarely saw spending decrease, especially after 1943. In the 1919-20 school year, adjusted for inflation, we spent $13.2 billion on public elementary and secondary education. In 08-09 we spent almost $690 billion, or about 52 times as much. Of course enrollment also grew—we only spent about 23 times as much per pupil!

What was the magnitude of the retrenchment between the peak spending year of 08-09 and the recession spending trough, 12-13? Total spending fell from about $690 billion to $636 billion, or 7.8 percent. The average per-pupil expenditure dropped from $13,816 to $12,621, an 8.6 percent decrease. Those dips aren’t nothing, but they are hardly catastrophic. And as of 14-15—the most recent year with federal data—total spending was back up to nearly $668 billion, and per-pupil spending to $13,119.

Teaching staff has also increased long-term. In previous posts we went back to 69-70 for data, but could have gone back to 1955 for national-level figures. Between that year and 2008, public schools went from about 3.7 teachers for every 100 students to 6.5, or about a 76 percent increase. That dropped just slightly—to 6.3 teachers per 100 students—in 2012.  

Teaching staff grew notably over the decades, but non-teaching staff growth has been far more remarkable. Going back to 1949-50, administrative staff per pupil has more than doubled, though still with only one administrator per 325 students. Support staff has grown more than three times larger, from 1 staffer per every 81 kids to about 1 per 26 students. Principals and assistant principals have more than doubled per-pupil.

Even with massive increases in resources, we haven’t seen inflation-adjusted teacher salaries increase all that much. Since 69-70—the farthest back federal data go—salaries have only risen about 6.4 percent. Total per-pupil expenditures, in contrast, grew by around 130 percent. What gives?

All that hiring, especially of non-teachers, for one thing. We’ve hired more teachers relative to enrollment, while teachers as a share of all public schooling staff dropped from over 70 percent in 1949-50 to just below 50 percent in fall 2015. Overall, in 1949-50 there were 19.3 students per staff member of all types—teachers, administrators, guidance counselors. In 2015 there were only 7.9. That’s a lot more salaries over which to spread money. We have also seen benefits’ share of compensation grow (see below). In 00-01 benefits accounted for 17 percent of total, current per-pupil expenditures in public schools, and salaries 64 percent. In 14-15 benefits had moved up to 23 percent and salaries 57 percent.

Short-Term & In Some States: A Historically Rare Case of (Some) Cuts

The Great Recession precipitated real cuts, something seldom seen in public schooling since the early 20th Century. But the cuts were limited, varied greatly by state, and did not occur in all areas of spending.

Between 99-00 and 14-15—the time period for which we could put together consistent, total state-level spending—outlays per-pupil nationally rose from $11,510 to $13,119, though they dropped from a peak of $13,816 in 08-09. Various services, meanwhile, saw increasing outlays not just through the whole period, but also after the recession. This is consistent with a very long-term trend: hiring more and more non-instructional staff, perhaps to deal with ever-increasing bureaucratic demands on schools, as well as assigning to schools increasing non-academic missions. The area where we saw the most significant drop in spending was capital outlays—buying land and erecting buildings.

How about specifically in the restive states? Spending cuts were not necessarily the name of the game. Arizona—which has seen huge increases in enrollment since 1969-70—increased overall spending between 99-00 and 14-15, with a big increase between 99-00 and 07-08. But that could not keep up with enrollment; on a per-pupil basis spending dropped over the period. Colorado also saw overall spending rise, but it just barely fell short of keeping per-pupil funding equal from the beginning to the end of the period. North Carolina increased overall spending slightly, but saw a decline per-pupil. Kentucky, Oklahoma, and West Virginia, in contrast, saw both overall and per-pupil spending increase.

In terms of what’s been trimmed, some buffeted states saw significant cuts in capital outlays, but that category of spending tended to be very volatile. Generally, states seemed to largely protect or even increase instructional spending, while all saw increases in spending for various types of services, a finding consistent with the increased administrative spending and staffing we have seen nationally. All except Kentucky and Oklahoma have had decreasing teacher salaries since 99-00, and every one of the hot-spot states has seen long-term stagnation in teacher salaries, which is roughly the national trend.


If someone tells you that public school spending has been “gutted” or “cut to the bone,” or any other body-destroying description, the first thing to note is that for many decades prior to the Great Recession we shoved so much food into the public schooling system that it would more accurately have been seen as threatened with obesity than “gutting.” Even since the recession, we haven’t typically gutted anything—significant funding has still flowed—and that includes in most embattled states. That said, at least based on salaries, teachers have seen their compensation stagnate. However, a lack of overall public schooling resources is not to blame for this. It is other things: huge increases in hiring of non-teachers, and compensation moving more toward benefits than salaries.

Farm Bill Flop

Last week, the House voted down passage of the 2018 farm bill. The bill would have reauthorized farm programs and food stamps at a 10-year cost of $867 billion. Democrats voted in a block against the bill because they opposed expanded work requirements for food stamps. A group of Republicans voted against it because they were frustrated by a lack of action on unrelated immigration legislation.

Few members objected that the bill was a budget buster, and would not have trimmed bloated spending on either farm or food subsidies. Federal deficits will soon top $1 trillion a year, and the projected growth in debt in coming years is unprecedented in American history. Subsidies of all sorts need to be cut.

We have two parties, and their members tend to fall on opposite sides of many issues, such as taxes, gun control, and abortion. Likewise, we would expect to have one pro-subsidy party and one anti-subsidy party. But we do not.

The alleged conservative party now in control of the government should be the anti-subsidy party, but big spenders dominate the Republican caucus. Politically, I think that is bad positioning for the GOP, as it leaves millions of fiscally conservative voters with nowhere to go. But we will see what happens in November.

Meanwhile, policymakers need to figure out next steps for the farm bill. I make suggestions in this new op-ed at The Hill.

For background on federal farm policies, see here.

While Politicians Cut Opioid Prescriptions, Fentanyl—With Help From the “Dark Web” and the USPS— Becomes the Number One Killer

A May 22 story in Bloomberg News describes with painstaking detail the underground pipeline through which the powerful synthetic opioid fentanyl floods the US market. According to the Drug Enforcement Administration, while the Mexican cartel plays a role by using its well-established heroin and methamphetamine distribution networks, most of the fentanyl comes in to the US from China. 

The raw materials to make the synthetic opioids are cheap and they can be manufactured rather quickly in small laboratories. The laboratories are constantly creating new variations so as to skirt restrictions the Chinese government places on existing fentanyl analogs. Online distributors throughout China sell these products, making their transactions over the “dark web,” often paid with cryptocurrency, and frequently ship the products to the US via the US Postal Service or United Parcel Service. 

Many dealers purchase and use pill presses to make counterfeit OxyContin or Vicodin pills and trick non-medical users into thinking they are buying the real thing. That’s how Prince died. He preferred to abuse Vicodin (hydrocodone). Records show he never got prescriptions from doctors. He died from ingesting counterfeit Vicodin pills he obtained on the black market that turned out to be fentanyl.

The DEA reports this is the way most fentanyl makes its way to the street. As we doctors know, most pharmaceutical-grade fentanyl made for medical use does not get diverted on to the streets. In fact, the forms usually prescribed to outpatients—skin patches, lozenges, buccal films—are not very suitable for non-medical use.

The Centers for Disease Control and Prevention reports that fentanyl was responsible for 26,000 overdose deaths in 2017. But already in 2016 fentanyl accounted for more than 20,000 of the roughly 64,000 total overdose deaths (which include cocaine, methamphetamine, and benzodiazepines). Heroin came in second with more than 15,000. In fact, for a few years now, fentanyl and heroin have accounted for the majority of overdose deaths. And a great majority of those deaths had multiple other drugs on board. In New York City in 2016, three-quarters of overdose deaths were from fentanyl and heroin, and 97 percent of overdoses had multiple other drugs on board—46 percent of the time it was cocaine.

Fentanyl overdoses in the US have been rising at a rate of 88 percent per year since 2013. Heroin overdoses have been increasing at a rate of 19 percent per year since 2014 after climbing 33 percent per year from 2010-2014. Meanwhile, overdose deaths from prescription-type opioids have been increasing at a stable rate of 3 percent per year since 2009.

The National Survey on Drug Use and Health reports non-medical use of prescription opioids peaked in 2012, and total prescription opioid use in 2014 was lower than in 2012. And the survey repeatedly reports less than 25 percent of non-medical users see a doctor in order to get a prescription. Three-quarters obtain their drugs through a friend or family member or a drug dealer.

Meanwhile, while all this is going on, policymakers in Washington and in state capitals seem intent on getting the opioid prescription rate down further. State-based prescription drug monitoring programs have succeeded in reducing the prescription of high-dose opioids by over 41 percent since 2010, the peak year of opioid prescribing. And opioid production quotas, set by the DEA, were reduced 25 percent last year and another 20 percent this year, generating acute shortages of injectables in hospitals across the nation that is harming patients.

With all the evidence that the majority of non-medical users are not patients—with all the evidence that prescription rates have come down while overdose rates keep going up—with all the evidence of fentanyl and heroin flooding the black market and causing those deaths, it is time for policymakers to disabuse themselves of the false narrative to which they’ve been stubbornly clinging. This narrative blames the overdose problem on doctors prescribing pain relievers to their patients. The overdose problem has always been primarily caused by non-medical users accessing drugs in the dangerous black market created by drug prohibition. And our current restrictive policy is only driving up the death rate by pushing these users to more dangerous drugs while making patients suffer in the process.

What’s the definition of insanity?

Mandatory E-Verify will Increase Identity Theft

Nancy Berryhill, an Acting Commissioner of Social Security, recently testified in front of the House Subcommittee on Social Security on the widespread use of Social Security Numbers (SSNs) beyond their intended function.  Most of her testimony concerned the history of SSNs, past security procedures, and proposed future ones.  In a bizarre sentence that contradicts much of the rest of her testimony, Berryhill stated that, “Mandatory use of E-Verify by employers would help reduce the incidence of fraudulent use of SSNs.”  That is exactly backward.  Mandatory E-Verify will greatly expand the fraudulent use of SSNs.

E-Verify is an electronic employment eligibility verification system run by the federal government that is supposed to check the identity information of new hires against government databases to verify that they are legally eligible to work.  Congress created E-Verify to deny employment to illegal immigrants and reduce the incentive for them to come and remain in the United States.  E-Verify is not yet mandated nationwide but several states have mandated its use, to various degrees, and many large employers currently use it.

E-Verify builds on the current rudimentary employment verification known as the I-9 form that every new employee must fill out thanks to the 1986 Immigration Reform and Control Act (IRCA).  An E-Verify mandate would add another layer on top of the I-9 whereby employers, after collecting I-9 forms, would enter the information on them into a government website.  The E-Verify system then compares that I-9 information with information held in the Social Security Administration (SSA) and Department of Homeland Security (DHS) databases.  The employee is work authorized if the databases decide that the information is valid.  A flag raised by either database returns a “tentative non-confirmation,” requiring the employee and employer to sort out whatever error has been flagged.  If the employee and employer cannot sort out the errors then the employer must terminate the new employee through a “final non-confirmation.”  The I-9 form and E-Verify have serious problems, including the encouragement of rampant identity theft, but those problems would only grow with an E-Verify mandate.

Epic Systems v. Lewis: It’s OK To Calm Down About Arbitration

Yesterday’s 5-4 Supreme Court decision upholding agreements to individually arbitrate wage-and-hour claims was neither surprising nor novel as a legal matter. Nor – notwithstanding the variously breathless, furious, and apocalyptic reactions it has drawn from stage Left – is it objectionable as a matter of policy, or “anti-worker.” It is pro-liberty, pro-contract, and pro-respect for private ordering.

On a practical level, the decision in Epic Systems v. Lewis and two companion cases leaves in place (rather than changing) a by now familiar business practice. Not until 2012 did the National Labor Relations Board embrace the notion that the National Labor Relations Act bans arbitration agreements requiring individual rather than class-action pursuit of wage claims, and that Obama-era position has not been able to maintain the assent even of the full federal government (the current Department of Justice disagrees, and filed against the agency’s position). 

At the level of legal precedent, this is by one count the seventh major Court decision since 1983 confirming (in each case over liberal dissents) that the Arbitration Act’s broad federal policy favoring arbitration agreements is compatible with, rather than ousted by, some other federal law. In fact, in one of those decisions, 1991’s Gilmer v. Interstate/Johnson Lane Corp., the Court had already implied (in interpreting a parallel statutory scheme) that the federal statute directly governing wage and hour suits, the Fair Labor Standards Act (FLSA), does not ban this kind of arbitration agreement.  That foray having yielded nought, advocates came back with a “bank shot” (Justice Neil Gorsuch’s phrase) theory that even if the FLSA doesn’t forestall individualized arbitration of FLSA claims, the National Labor Relations Act does, under a miscellaneous clause that extends federal legal protection to some “concerted” activity by employees that does not consist of union action. But although some labor movement advocates have hoped to use this catchall language as the future engine by which the NLRB would gain power to impose major new regulatory requirements at non-union workplaces – all sorts of on- and off-job interactions between colleagues might be interpreted as concerted activity if you squint at them the right way – it was always doubtful that the current Court would go along with a very broad reading on that. 

Chicago Police Don’t Need Facial Recognition Drones

Chicago Mayor Rahm Emanuel is backing legislation that passed the state Senate earlier this month that would allow Illinois police to use drones to monitor “large scale events,” including protests. This legislation would be worrying enough if the drones were merely outfitted with video and audio capability. However, these drones could one day be equipped with facial recognition tools, amplifying the privacy risks associated with drones buzzing over citizens engaging in First Amendment-protected activities.

Supporters of drone surveillance such as State Senator Martin Sandoval (D-11th District) cite public safety concerns as justification for this bill. But public safety can and is cited for any new piece of surveillance equipment. When considering the deployment of surveillance technology we should consider how the technology is likely to be used, not how its proponents say it will.

The proposal, backed by two of Emanuel’s General Assembly allies, is an amendment to Illinois’ Freedom from Drone Surveillance Act, which includes some admirable provisions, such as a warrant requirement. If passed, police would be permitted to use drones to surveil any event with at least one hundred people in attendance. Protests and demonstrations are only a few of the events that could fall into this category – football games, parades, music performances, and conventions would also be fair game for drone surveillance. 

Chicago police are already technology pioneers, taking advantage of what University of the District of Columbia law professor Andrew Guthrie Ferguson calls “Big Data Policing.” In Chicago, police use a secret algorithm that assigns a police risk score to hundreds of thousands of residents. Tens of thousands of these residents are classified as “high risk” of being involved in a shooting despite having never been arrested or shot. 

The Chicago Police Department has been criticized for conducting social media surveillance, and a few years ago it acknowledged that it had been using cell-site simulators – powerful snooping tools originally designed for military use. Given the CPD’s propensity for new surveillance gadgets we should expect its officers to fly drones over protests and similar gatherings if provided the opportunity.