Filter | only charts

Common Core Goes to College

Christina Breitbeil - July 31, 2014

In a new paper, the New America Foundation's Lindsey Tepe writes that Common Core assessments should be used in college admissions. We recently spoke with Tepe to learn more; the conversation has been edited for length and clarity.


What is Common Core, and how do you think its implementation will affect college admissions?

The Common Core State Standards are a set of state standards from kindergarten to twelfth grade that inform what content children need to master by the end of each grade, culminating in college and career readiness by the end of twelfth grade. Part of that would also be the Common Core assessments, which are being developed by the PARCC [Partnership for the Assessment of Readiness for College and Careers] and Smarter Balanced Assessment Consortia, and those assessments will track students' progress of mastery of those standards through each grade.

So, right now they may not affect college admissions. But these standards were developed with the idea that they are college- and career-readiness standards. Other states that have not adopted the Common Core State Standards have adopted different college- and career-readiness standards, and so what all of these standards have in common is that they have said that if students master these particular skills and this particular set of information that's embedded in them, that they'll be ready to go on to college. But that's very different than saying that students will be admitted to college.

College admissions are decided on an institutional basis, so each college and university has their own admissions criteria, some of which are open enrollment -- like community colleges -- and all you need is a high-school degree for the most part. Other selective-enrollment institutions use a combination of test scores, high-school GPA, and other factors, as they determine which students will be admitted.

Looking forward, however, if states have made a commitment to college- and career-readiness standards and assessments, it seems very silly that they would not also then say that if you meet your state's standard, that should then qualify as a minimum admissions standard to the lowest tier of four-year universities in that state.


Can these new assessments replace the ACT and SAT completely, then?

I would not say that they would replace, or that they necessarily should replace, the ACT or SAT. What I am saying, though, is that in many cases, universities have a minimum ACT and/or SAT score that you must meet to gain admission. Now, if a state has adopted the Common Core State Standards and they have adopted either the PARCC or the Smarter Balanced Assessment, and if you are scoring college-ready on those tests, I argue that it does not make sense for a student to have to both pass college-ready on PARCC or Smarter Balanced as well as college-ready on ACT or SAT in order to gain minimum admission to a university.


How do you think policy should be shaped to best facilitate the implementation of these standards?

We don't want this effort to stop at the college door. In K-12, curriculum decisions are increasingly being made at the state level, as opposed to the local level. There is a prescribed list of things that local school districts can purchase, for example. And at the federal level, with No Child Left Behind, you have a whole host of requirements geared toward quality of schools, and regardless of how you feel about that, states are really looking to improve quality in K-12 education and looking at outcome. In higher education, you just don't see that kind of scrutiny, and with the push toward college completion, with looking at graduation rates, what we're seeing is that higher education often just doesn't have the same level of attention paid to what they're actually doing for the students who are enrolled there.

What I think could really be done is to use the Common Core or the college- and career-readiness standards that the state has adopted to then look and see how students, being educated to a very high level up through twelfth grade, are doing in higher education, and start to put more accountability for outcomes in institutions of higher education.


In the paper, you say that Common Core standards and aligned assessments will cause "educational data [to] grow more comparable at the national level." Why do we need this?

When No Child Left Behind first passed, and states started implementing their own individual assessments, they were allowed to design their own tests, and they were allowed to set the bar for proficiency at whatever level they wanted. With these tests, you had all these states saying, "Our students are proficient; our students are doing really well." Yet, when they took a national assessment like NAEP [the National Assessment of Educational Progress], you were suddenly seeing that those state tests were not measuring students very well. You had states like Louisiana that were testing in the 80th percentile for proficiency, and on the national assessment, it was more like 40 percent.

When you have similar students from different states being told very different things about how proficient they are in math and reading, we have a little bit of a problem, because students don't always stay in their same town through their whole career. They might move to a different state; they might attend a university in a neighboring state; they might have a job out of state. Even if they stay in the same location, they may have students from other states coming to their university, or they might have other students from across the country moving to compete for jobs in their same area.

So, having students receive different messages about their proficiency really starts to show up when they get to college, or when they get to their career, and suddenly they realize they don't have the same skills as the people they are working with, or that they were not receiving the best information about their level of preparedness. It's the same reason why, when students from other countries are looking to come to the United States to go to a university, they take the ACT or SAT. Institutions want to know based on their test, based on those tools, how students compare to the population that's already here.


Christina Breitbeil is a RealClearPolitics editorial intern.

Robert VerBruggen - July 31, 2014

Over at Investor's Business Daily, Jed Graham has some interesting data on the ratio of people working 31-34 hours per week to those working 25-29 hours:

Under Obamacare, starting in 2015, many employers will be required to provide health coverage to employees who work 30 hours or more. Anecdotal reports of employers cutting hours in anticipation are plentiful, and last year the White House itself used this metric to argue that the law wasn't hurting the workweek.

Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen

Progress on Biosimilars: Not Fast Enough

Alan Daley - July 31, 2014

Drugs that compete with brand-name drugs are usually called "generics." Off-brand products that compete with an especially complex category of drugs called "biologics," however, are called "biosimilars." Whereas generics are chemically identical to brand-name drugs, the FDA defines a biosimilar as having a new active ingredient while being highly similar to the reference product and exhibiting no clinically meaningful differences in safety, purity, and potency.

Examples of biologics include Humira (for rheumatoid arthritis), Epogen (for anemia due to chemotherapy), and Herceptin (for breast cancer), and such drugs are usually priced at high levels. For example, Humira is likely to cost a patient $50,000 per year, and some drugs for rare diseases can cost nearly $500,000 per year. Congress foresaw the importance of fostering biologic competition in 2009 when it passed the Biologics Price Competition and Innovation Act.

Certainly, the FDA needs to ensure that each biosimilar is as safe and effective as its biologic predecessor. Unfortunately, the FDA has not yet moved on to the stage of judging safety and efficacy, because it seems to be stuck on administrative issues such as a naming protocol for biosimilars and a suitable FDA fee structure for anyone who might venture to compete in the biosimilar space.

Although the FDA is unready to approve biosimilars, its regulatory counterparts in the European Union, Canada, Australia, and some Asian countries have already approved the manufacturing and dispensing of biosimilar drugs. Clearly, there are some pharmaceutical manufacturers who are able to produce safe and effective biosimilars. In the countries where biosimilars are allowed to compete, prices are already about 20 percent lower, and the CBO predicts that U.S. biologics' prices will be 40 percent lower after four years of biosimilar competition.

One of the more successful biologic drugs, erythropoietin, produces $100 million per month in profit. This means that aggressive competition for that one drug could yield up to $100 million per month in consumer savings. The FDA's failure to progress on biosimilars has a massive financial impact on consumers.

Express Scripts estimates a $250 billion savings for consumers over the next eleven years if competition by biosimilars is allowed. This means that, over the next 11 years, FDA inaction on biosimilars will average $2,000 in costs per household in the U.S.

Some in the pharmaceutical industry would rather talk about drug naming. There is a dispute over whether each biosimilar, because it is chemically slightly different from the original, should have a unique name. To some, it may seem obvious that precise chemical names should be kept distinct (and the drug's market names can be subjected to our customary trademark laws). The resulting names won't be less consumer-friendly than those in use today -- pharmaceutical makers already abandoned using pronounceable names when it comes to chemicals as opposed to brands. Try saying adalimumab or erythropoietin or trastuzumab at a normal speaking pace.

On the other hand, a number of pharmacy associations have concluded that distinct names would cause confusion and possibly medical errors. Whatever the resolution, the naming convention can be decided after the FDA establishes a pathway for competitive entry and while biosimilars make their way through the regulatory pipeline to final approval. That process will allow for these disputes to be settled.

The fact is that the naming issues have become a time-consuming task for the FDA. Shame on the FDA for allowing itself to be sidetracked by this nonsense -- it's a red herring kept alive by those benefiting from highly profitable biologicals. It should not delay getting safe products to market and justify forcing consumers to overpay. It has been happening in Europe since 2006, and the delay at the FDA is totally unwarranted.

Inaction on approving biosimilars harms consumers and benefits biologics patent-holders. The FDA seems unduly tolerant of its own slow progress, and this needs to change.

Alan Daley is a retired businessman who writes for the American Consumer Institute Center for Citizen Research.

Quantifying the Cost of NSA Snooping

Steven Titch, R Street - July 30, 2014

In the year since Edward Snowden began disclosing the scope of National Security Agency’s programs to use cell phone networks, the Internet and various commercial websites to spy on both American citizens and foreign nationals, there has been considerable speculation about the cost of these programs to the U.S. information technology industry in terms of money and trust.

The New America Foundation has released a report that attempts to quantify these costs, concluding that over the past 12 months the NSA’s actions “have already begun to, and will continue to, cause significant damage to the interests of the United States and the global Internet community.”

In the executive summary, authors Danielle Kehl, Kevin Bankston, Robyn Greene and Robert Morgus discuss detrimental effects on four specific areas:Direct economic costs to U.S. businesses: American companies have reported declining sales overseas and lost business opportunities, especially as foreign companies turn claims of products that can protect users from NSA spying into a competitive advantage. The cloud computing industry is particularly vulnerable and could lose billions of dollars in the next three to five years as a result of NSA surveillance.

• Potential costs to U.S. businesses and to the openness of the Internet from the rise of data localization and data protection proposals: New proposals from foreign governments looking to implement data localization requirements or much stronger data protection laws could compound economic losses in the long term. These proposals could also force changes to the architecture of the global network itself, threatening free expression and privacy if they are implemented.

• Costs to U.S. foreign policy: Loss of credibility for the U.S. Internet freedom agenda, as well as damage to broader bilateral and multilateral relations, threaten U.S. foreign policy interests. Revelations about the extent of NSA surveillance have already colored a number of critical interactions with nations such as Germany and Brazil in the past year.

• Costs to cybersecurity: The NSA has done serious damage to Internet security through its weakening of key encryption standards, insertion of surveillance backdoors into widely-used hardware and software products, stockpiling rather than responsibly disclosing information about software security vulnerabilities and a variety of offensive hacking operations undermining the overall security of the global Internet.

Among the recommendations the authors make are strengthening privacy protections for both Americans and non-Americans, within and outside the U.S. borders; increased transparency around government surveillance, both from the government and companies; renewed commitment to the Internet freedom agenda in a way that directly addresses issues raised by NSA surveillance, including moving toward international human rights-based standards on surveillance; and development of clear policies about whether, when and under what legal standards it is permissible for the government to secretly install malware on a computer or in a network.

The entire 64-page report can be downloaded here. A summary version is here. Also check out my own R Street white paper on the potential cost of NSA computer spying here.

Steven Titch is an associate fellow of the R Street Institute. This piece originally appeared on R Street's blog.

More Immigration, Stronger Economy

Jason Russell - July 30, 2014

Debate rages in Congress over what to do with the tens of thousands of unaccompanied children who are trying to join their relatives in America. There can be no clearer sign of the need for immigration reform than children putting themselves in physical peril for a better life in the U.S. Increased legal immigration and a simpler, shorter process for crossing the border would have enormous humanitarian benefits.

But these reforms would also help solve America's economic and budgetary problems. The weak recovery and an aging population have labor-force participation hovering around its lowest rate in 35 years. America's birth rate is less than half of what it was a century ago. The cost of programs such as Medicare and Social Security is projected to raise publicly held federal debt to over 100 percent of GDP by 2036. And there is good evidence that immigration could help to address all of this.

Research by Giovanni Peri at the University of California, Davis, shows that more immigration will raise U.S. wages and create additional jobs for native-born Americans. Immigrants bring different skills into the economy and typically do not compete for the same jobs as natives. Further, immigrants are generally paid lower wages, which allows businesses to expand and hire additional workers.

Beyond the benefits to the economy as a whole, all this adds up to more tax revenues. The Congressional Budget Office has concluded that 16 million additional immigrants by 2033 would decrease the federal budget deficit by $700 billion over 2024-2033 period.

Some worry that immigration will strain the safety net, but low-income immigrant households with children rely less on the safety net than do similarly-situated native households. Welfare reform in 1996 eliminated most eligibility for immigrants -- even legal immigrants, who had previously been equal to U.S. citizens under welfare law. As a result, immigrants have a labor-force participation rate almost two percentage points higher than native-born Americans. The vast majority of immigrants come here to work, not to take advantage of our welfare system.

Immigrants do impose some economic costs in the short term. Immigrants tend to have more children than native-born Americans, putting pressure on school districts. U.S.-born children of immigrants are eligible for welfare programs, regardless of their parents. And immigrants have a low rate of health-insurance coverage, burdening health providers.

However, the longer-term gains of increased immigration outweigh these costs. A 2012 literature review by Daniel Griswold shows that the largest factors behind increased K-12 spending are special education and administrative overhead, which have nothing to do with immigration. Griswold cites the 2011 Social Security Trustees Report, which says 300,000 more immigrants per year would extend Social Security's solvency by one year. Griswold also shows that, despite having a lower health-coverage rate than native-born Americans, immigrants are younger and healthier and do not significantly burden the overall health-care system.

As the children of immigrants grow up, they will begin paying taxes and contributing to the economy -- most likely, some will even become great entrepreneurs and scientists. For instance, Mario Capecchi came to the United States in 1946 as a nine-year-old. He went on to earn a Ph.D. in biophysics from Harvard and won the Nobel Prize in 2007. Elon Musk came to the United States during college and later co-founded PayPal, SpaceX, and Tesla Motors. The next great technological breakthrough may well come from a foreigner, and it is in our interest to have that individual here, not brushed aside by overbearing immigration rules.

Today, America has the world's strongest economy, and immigrants will help ensure that it remains that way. Immigrants and their children should not have to risk their lives to come to the land of opportunity.

Jason Russell is a research associate at Economics21, a project of the Manhattan Institute.

Is Detroit's Pension Deal a Game-Changer?

Richard Johnson, Urban Institute - July 29, 2014

Detroit's municipal employees and retirees made headlines last week when they accepted a pension cut to help reduce the city's debt. Is this agreement a game-changer, as some contend? Now that some retirees have agreed to forgo already earned pension benefits -- once considered sacrosanct -- will unions across the country follow suit? Don't count on it.

Many state and local employees and retirees have already shouldered significant pension cuts, and it's unrealistic to expect them to absorb any more. Rather than relying on givebacks from public servants, policymakers must commit themselves to paying the benefits they promised.

Like most public-sector employees, city workers in Detroit who spent their entire careers on the city's payroll receive generous pensions. Benefits equal a fraction of employees' final average salaries multiplied by years of service. Before the city's finances collapsed, that fraction rose over employees' careers, reaching 2.2 percent of salary after 25 years on the job. Employees with 30 years of service could retire as early as age 55. Once retired, their benefits automatically rose 2.25 percent a year.

Over a lifetime, these benefits add up. Consider a Detroit municipal employee making $50,000 a year. Under the old rules, after 30 years on the job, he could retire at the age of 55 with an initial pension of $27,500, worth more than half a million dollars over his lifetime (assuming a 2 percent real interest rate and 3 percent inflation).

As the city prepares to enter bankruptcy, however, that pension has been slashed. First, the city cut the benefit-formula multiplier for years worked after 2011 to just 1.5 percent. That change trimmed lifetime benefits for newly hired city workers by about a fifth. Last week's agreement further reduces annual benefits by 4.5 percent and eliminates the automatic benefit escalator in retirement, reducing lifetime benefits for new retirees by another quarter. New hires who eventually retire at age 55 after 30 years of service will receive pensions worth 40 percent less, in inflation-adjusted dollars, than their counterparts who retired in 2011.


New Jersey Employees Have Already Taken on the Burden of Benefit Cuts
These cuts may help get Detroit back on its feet, but don't expect similar retiree givebacks elsewhere to solve the nation's public pension problem. Many state and local retirement plans have already substantially cut benefits. Most of the benefit formula changes apply only to new hires, but current workers and retirees have not been spared.

As the Center for Retirement Research points out, between 2010 and 2013, 12 states reduced or eliminated cost-of-living adjustments for current retirees as well as current employees and new hires. (Another five states reduced COLAs but shielded current retirees). Many jurisdictions have also raised the amount employees must contribute to their plans.

New Jersey, with some of the worst-funded plans in the nation, is a good example. In 2011, lawmakers eliminated COLAs for state retirees, whose benefits had increased each year by 60 percent of the change in the consumer price index. That cut shaved 25 percent off the lifetime pension of a newly retired state employee with 35 years of service. The state also increased mandatory employee contributions from 5.5 to 7.5 percent of pay, reducing what retirees get from the state by another eighth. Additionally, the 2011 reforms boosted the retirement age and reduced the plan multiplier.

All told, New Jersey state retirees will receive pensions only two-fifths as large as what they would have been paid under the rules in effect before 2008. Yet, New Jersey's pension plan is in worse financial shape today than it was in 2007. Because of these benefit cuts, state employees hired at age 25 must now work 28 years before their future payments are worth more than the value of their required plan contributions. Those who leave state employment earlier end up financing their entire pensions themselves, without any state contributions. In fact, they would be better off if they could opt out of the state retirement plan and invest their required contributions elsewhere.


States Should Fully Fund Their Retirement Promises
Most troubled public pensions are financially distressed because states and localities have not contributed as much as their actuaries say they must in order to pay the future benefits they've promised, not because their benefits are too generous. Our recent comprehensive analysis of state plans graded many plans poorly because they failed to provide employees with adequate retirement security, especially those who spent less than a full career in public service. If policymakers want to fix the public pension mess, they must dedicate themselves to fully funding the retirement promises they've already made.


Richard Johnson is a senior fellow with the Urban Institute and director of the organization's 
Program on Retirement Policy. This piece originally appeared on the Urban Institute's MetroTrends blog.

Higher-Ed Accreditors Have Too Much Power

Michael Poliakoff - July 29, 2014

Every year, the federal government distributes money to colleges, universities, and trade schools across the country in the form of subsidized loans and grants. Federal education funding is essential to these institutions -- losing it is often a death sentence. But under the Higher Education Act, decisions about which schools can and cannot receive government money rest not with Congress or the secretary of education, but with private accreditation agencies that the secretary of education deems reliable. The groups are largely composed of education-industry stakeholders -- and they are given wide latitude not only to enforce, but also to set, the standards that determine which schools are eligible for federal funds.

These accreditors exercise their authority in an increasingly arbitrary manner, failing to ensure educational quality while obstructing effective university governance and educational innovation. It is now clearer than ever that the federal government's delegation of gatekeeping power to accreditation agencies is bad public policy.

It may also be unconstitutional.

On July 17, the American Council of Trustees and Alumni (of which I am vice president of policy), along with the John W. Pope Center for Higher Education Policy and the Judicial Education Project, filed an amicus brief in the Fourth Circuit Court of Appeals in the case of Professional Massage Training Center Inc. v. Accreditation Alliance of Career Schools and Colleges. We argue that the delegation of government authority to a private party with a conflict of interest is grossly at variance with legal precedent and jeopardizes the political accountability that the Constitution is designed to safeguard.

For nearly 80 years, the Supreme Court has restricted Congress's ability to delegate the powers of government to private parties. This kind of statutory delegation is valid only if the role of the private party acts in a "ministerial or advisory" capacity, not if it is given "core governmental power." The latter better describes the role of accreditors: Like a government agency, they enforce rules that they themselves have made. And surely the accreditors' ad-lib standard-setting and gleeful intrusion into matters rightfully under the purview of state government or statutorily constituted governing boards meet and exceed the definition of a "core governmental power."

Accreditation requirements exist in order to ensure that postsecondary institutions receiving taxpayer money meet basic standards of quality, but accreditors too often fail to achieve this simple goal. Many college graduates cannot reliably answer questions that require the comparison of viewpoints in two editorials, or compute the cost per ounce of food items. Richard Arum, Josipa Roksa, and Esther Cho have demonstrated that many students graduate college without experiencing significant cognitive gains. These facts become far less surprising when one considers that many accrediting bodies' standards do not make any attempt to deal directly with student learning outcomes.

Accreditors do seem to find time to interfere with the internal governance of the schools they accredit, though. The Southern Association of Colleges and Schools Commission of Colleges placed the University of Virginia on "warning" status because the association disapproved of the process by which the school's governing board attempted to fire the university's president. The Accrediting Commission for Community and Junior Colleges revoked the City College of San Francisco's accreditation largely on the basis of failures to reform matters of administration and governance.

As the law stands, there is virtually nothing that can be done about this. Under the statute, the accreditors approved by the secretary of education have the sole authority to set standards. And without a legislative change, the judiciary remains the only authority with the power to oversee this arrangement.

But as vital as judicial review is to prevent the misuse and overreach of delegated federal authority, true reform must come from Congress. Only when accreditors return to their role as voluntary peer reviewers, and the government starts focusing on creating more transparency and choice, will we begin to see real change in higher education.

Michael Poliakoff is vice president of policy at the American Council of Trustees and Alumni.

Does Economics 101 Apply to Immigration?

Robert VerBruggen - July 29, 2014

It's basic economics: When supply expands, prices fall. So one might think that when immigrants increase the supply of labor, the price of labor -- that is, employee compensation -- will decline. But it's "suprisingly difficult" to demonstrate that this actually happens, according to the famed Harvard labor economist George Borjas.

Borjas's new book, Immigration Economics, spans a mere 215 pages (excluding appendices and notes), and yet it stands as a testament to the immensity of this challenge. Reading Borjas -- or, more accurately, wading through his intricately woven thicket of economic models and the theoretical justifications for them -- one can't help but suspect that the entire enterprise is doomed.

While discussing one approach to measuring immigration's impact on wages, for instance, Borjas notes that different ways of grouping people by education level "can make the estimated wage effects as small or as large as one would like." Discussing another method, he shows that focusing on different types of geographical areas -- states, metropolitan areas, etc. -- can also dramatically change the results, for reasons we don't fully understand. (Part of it seems to be that immigration to a metropolitan area causes natives to move elsewhere, spreading out the impact.) He even reveals flaws in the basic datasets that national governments make available to researchers.

Given the malleability of the numbers and the politically charged nature of the topic, it should hardly be a surprise that experts disagree. For his part, Borjas has for years been arguing that the Economics 101 story has a lot of truth to it -- immigration is certainly good in many ways, but it drives down wages for competing native workers. Others, however, argue that immigrant workers "complement" natives instead of competing with them, and thus higher immigration will improve the wellbeing of just about everyone.

This book won't be the last word, obviously. But it does serve as a terrific introduction to the technical side of this topic -- at least for those willing to cope with multiple equations per page -- as well as a careful argument that immigration may not be all it's cracked up to be.


Studying immigration's effect on wages might seem simple: When immigrants come to a new place, do natives' earnings rise or fall? But there's much more to it than that. Immigrants do not choose their destinations at random -- they specifically head to places with jobs suited for their skills -- so it's very difficult to compare high-immigration and low-immigration areas. And as already mentioned, researchers need to make important decisions about how they sort their data into geographical regions and educational groups. A poorly conducted study can fail to pick up on key phenomena.

In decades of well-regarded research summarized in Immigration Economics, however, Borjas has built an empirical case that there are "distributional consequences" to immigration -- newcomers reduce the wages of the natives they compete with directly, but make others better off. In the end, the overall economic benefits of immigration to natives are quite small: In the various models presented here, a 15 percent expansion of the labor supply can't provide a "surplus" for natives amounting to even 1 percent of GDP. And high-school dropouts -- about 10 percent of the native U.S. population and 30 percent of immigrants -- often take a hit to their wages of several percent.

Does the complementarity argument fare better for higher-skilled workers? There are reasons to think it might -- even if you find the idea that immigrant high-school dropouts "complement" native high-school dropouts somewhat absurd, it's hard to deny that highly skilled workers often learn from each other. But Borjas presents some important data that undermines this notion as well, much of it drawn from academia, where the benefits of smart peers should be especially pronounced.

For example, analyzing supply shocks in various doctoral fields, Borjas concludes that a 1 percent increase in the supply of labor reduces earnings by 0.24 percent. Fluctuations in the cap on H-1B visas, meanwhile, in one study didn't seem to help (or hurt) high-skilled native workers in the metro areas where the immigrants went. When mathematicians left the Soviet Union in droves, American mathematicians in Soviet-dominated fields were able to publish less. And when Nazi Germany expelled Jewish academics from its universities, their departure harmed their students' careers but not their peers'.

As Borjas notes, this doesn't mean there are no benefits to complementarity. It just suggests that, from the perspective of an individual facing an influx of peers, the benefits don't outweigh (and often don't equal) the costs of new competition. Even if Borjas is correct here, the case for high-skill immigration remains pretty strong -- it's hard to be too upset when wealthy people encounter stiffer competition and as a result offer their services to the rest of us at a lower price. But if wage-boosting "complementarity" is hard to detect among the highly skilled, we should be rather skeptical of claims that it benefits the poor.


Of course, there's another side to the coin: benefits to immigrants. People disagree as to what role these benefits should play in making immigration policy, but hardly anyone would say they are entirely irrelevant. And considering that much of humankind suffers a level of poverty unknown in the First World, the gains offered by liberal immigration laws might be substantial indeed. Some economists consider open borders a "trillion dollar bill" for humanity just sitting there on the sidewalk.

The problem is that human beings tend to love the place where they grew up, and to establish networks of friends there. Borjas notes that when Puerto Rico became part of the United States, only a third of the population actually migrated, despite the huge economic opportunity available. This suggests that the psychic cost of uprooting oneself is quite high -- for most Puerto Ricans, the substantial benefits of moving to America weren't even worth the cost, and the Puerto Ricans who did move presumably benefited far less than the raw increase in their wages would suggest.

Studies within the U.S. find that moving from one state to another can cost hundreds of thousands of dollars once all this is taken into account. Borjas calculates, using some figures he concedes are at best ballpark estimates, that if we could magically move everyone in the world to the economically optimal place, these psychic costs might well outweigh the benefits. (He includes the present value of future generations' benefits through the use of a discount rate.)

The economist Bryan Caplan, a strong open-borders advocate, has offered a serious critique of this particular exercise. Most importantly, no one is advocating forcing people to move, and the people who willingly chose to immigrate under an open-borders regime would almost by definition be people who would profit from it.

But Borjas makes some less speculative points about the fate of immigrants in America as well, complicating the rosy narrative that open-borders advocates often present. One significant problem seems to be that "economic assimilation," the tendency of immigrants to start out behind natives but steadily catch up as they remain in the country, has slowed in recent decades. Immigrants' children often fail to catch up as well, thanks in part to segregated neighborhoods that transmit "ethnic capital" to each new cohort, keeping some ethnic groups from reaping the full benefits of immigration.

This clustering of ethnic groups in specific neighborhoods, which are continually replenished with fresh immigrants, also raises questions about balkanization. In a note, Borjas presents evidence that English acquisition among immigrants is slowing along with their overall economic assimilation.

And that brings us to the even bigger issue, which Borjas brings up at various points but is never able to address fully: At what point does immigration defeat its own purpose, recreating the problems immigrants are fleeing in a new place? As Borjas writes of fully open borders, it's a big assumption indeed that "whatever it is that makes workers much more productive in the developed countries [will remain] intact after the influx of hundreds of millions, if not billions, of immigrants." Regardless of how you parcel out the blame for the poverty of many nations, at least some relevant factors, most prominently culture, can travel right along with immigrants when they leave.

Borjas isn't being alarmist when he suggests that open borders could move an incredibly large number of people, either. Again, Puerto Rico lost only a third of its population when its residents gained the option of moving to the U.S. -- but a whole lot of people live in poor countries. Only about 15 percent of the world's population lives in developed nations, and billions of people are estimated to live on a few dollars a day.

Looking just at the Americas, whose residents would be most likely to head to the U.S. (total population 314 million, with 10 million illegal immigrants already), one-third of Mexico is 40 million people; Central America, 14 million; South America, around 100 million. It's not unreasonable to fear that with enough immigration, coupled with low levels of assimilation, America and other First World nations could become merely an amalgam, both economically and culturally, of the countries that so many were trying to escape. And the neighborhood trends Borjas has studied are not an encouraging sign.


Where does that leave us policy-wise? One shouldn't make too much of all this, considering that Borjas's findings are entirely open to dispute -- and that arguments about open borders are entirely theoretical. And even if one accepts Borjas's version of the facts, it's not too difficult to argue that the potential benefits of mass immigration outweigh the costs -- especially if you take a cosmopolitan approach, weighting benefits to natives and immigrants equally, and downplay the risks of balkanization and ethnic tensions.

But at the very least, Borjas provides the key elements of an argument against mass low-skill immigration: He lays out evidence that it undermines the employment prospects of the Americans who need decent jobs the most, and that it comes with risks of broader negative consequences.


Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen

Pitfalls Emerge in Health Insurance Renewals

Michael Ollove, Stateline - July 28, 2014

For the 8 million people who persevered through all the software trapdoors in the new health insurance exchanges and managed to sign up for coverage in 2014, their policies will probably automatically renew come November when open enrollment begins.

Seems like good news after all the headaches consumers endured after the program's launch last year. Except that renewing the same policy may not be the best choice. Many may end up paying far more than they need to and with policies that don't best fit their individual circumstances.

"(Automatic re-enrollment) could conceivably mean people will pay more in premiums unless they proactively take steps to comparison shop," said Jenna Stento, a senior manager at Avalere Health, a health care research and consulting firm.

If you made a good choice last year, what could be so wrong about re-upping with the same plan?

Turns out plenty, particularly for those among the 87 percent of enrollees in health insurance exchange plans who received a federal subsidy to help pay for premiums. Understanding why that's a problem isn't easy, the result of complicated quirks in the Affordable Care Act, which established the exchanges in the first place.


Premiums Up 8 Percent
Overall, premiums on the exchanges in 2015 may be a bit higher for most people, at least according to one analysis of proposed plans and rates in nine states. Avalere found that the average premiums for Silver plans will climb an average of 8 percent. (There are four grades of plans offered, starting with Bronze plans with the cheapest premiums, but higher deductibles and co-pays, and moving up to Silver, Gold and Platinum.)

The Obama Administration announced last month that consumers who bought their policies on the federal exchange would have them automatically renewed, as well as the amount of their subsidies.  It will be up to each state exchange whether to offer a similar automatic renewal. People whose level of income has changed would need to enroll again since it would affect the amount of their subsidies.

But consumers who automatically re-up with the plan they already have could face steep and unexpected premiums and out-of-pocket costs, particularly if they received a federal subsidy.


Changing Benchmark Plans
Here's why. The subsidy people receive is pegged to the second-lowest priced Silver plan, the so-called "benchmark plan," meaning that the amount of a subsidy any individual receives no matter which plan he or she selects, is based on how much they would receive if they picked that benchmark plan.

In a hypothetical example Avalere provides, "Sue," a Maryland resident, enrolled in the 2014 benchmark Silver plan in her region – offered by CareFirst Blue Cross -- which had a monthly premium of $214. Based on her income, Sue's contribution toward her monthly premium was set at $58, so she qualified for a monthly federal subsidy of $156 to make up the difference. If Sue had chosen a plan with a higher premium, her federal subsidy would have remained fixed at $156 and she would have had to pay more out of her own pocket.

However, in 2015, according to Avalere's analysis of early rate filings, CareFirst Blue Cross will no longer be the second lowest Silver plan in Sue's region but the ninth lowest out of 18 Silver plans, meaning that it will lose its status as the benchmark plan. CareFirst's new monthly premium is $267. The new benchmark Silver plan (the Silver plan with the second lowest premium) will be the Kaiser Foundation Health Plan with a monthly premium of $231.

Sue's contribution remains the same, but she will now qualify for a higher federal subsidy of $173 to make up the difference between her ability to pay $58 per month and the higher $231 monthly premium of the new benchmark.

If she automatically re-enrolls with CareFirst, however, she will have to cough up another $36 a month. By doing nothing, her out-of-pocket contribution will rise by 62 percent.

In another example, "Dave" enrolled in the benchmark Silver plan in Washington state, Group Health Cooperative, which had a monthly premium of $281. He received a federal subsidy of $85 each month, leaving him with a monthly out-of-pocket bill of $196.

In 2015, BridgeSpan Health will replace Group Health as the benchmark plan in Dave's area, with a premium of $263 a month. Because of that lower premium, Dave will be entitled to only a $67 a month federal subsidy, leaving him again with a $196 monthly out-of-pocket expense if he switched to BridgeSpan. But if Dave sticks with Group Health, which hiked its premiums to $313, he will have to pay $246 each month out of his own pocket, a nearly $600 increase compared to last year.

This is not a theoretical wrinkle. Of the nine states whose 2015 premiums Avalere examined (Connecticut, Indiana, Maryland, Maine, Oregon, Rhode Island, Vermont, Virginia and Washington), all but Vermont appear headed for a new benchmark plan when open enrollment commences. Consumers who live in six of these states may have an unpleasant surprise when they see their bills if they let their policies automatically renew.

In Rhode Island and Virginia, the opposite may be true. Last year's benchmark plans are expected to become the lowest price Silver plans, instead of the second lowest. Consumers renewing the 2014 benchmark plans in those two states could actually see their out-of-pocket premium costs decrease in 2015.

"There could be significant financial value to take a look at the site and see if there might be more affordable options for you, given the changes since last year," Steno said.


Website Tools
As re-enrollment approaches, numerous health care advocacy organizations, including Easter Seals, the March of Dimes, the Livestrong Foundation, the National Alliance on Mental Illness, and many others have urged the U.S. Department of Health and Human Services, which operates the federal health exchange, and the states that run their own exchanges to develop tools on their websites that will help consumers identify the plans that best fit their particular circumstances, not only in terms or premium costs, but also their actual usage.

In the first year, all exchanges showed the differences in premiums of the various health care plans as well as their differing cost-sharing formulas. Cost-sharing refers to deductibles, copays and co-insurance. (Copays are a fixed amount you pay for a particular medical service, such as $40 per primary care visit; co-insurance is a percentage that you have to pay for each service, such as 20 percent of a hospitalization.)

The lower the premiums, the higher the cost-sharing burdens on patients. As a result, cost-sharing formulas can result in the difference of thousands of dollars between one plan and the next, depending on an individual's or family's specific health care needs.

Those with chronic conditions, for example, who need many doctor visits in the course of a year, would do best to enroll in a higher premium plan with lower co-pays for individual visits. Relatively healthy people, on the other hand, would likely come out ahead by enrolling in a lower premium plan with higher co-pays.

That is why health advocates want all the exchanges to offer calculating tools that would enable customers to plug in information on their actual health care usage from the previous year to get an idea of how much they would be likely to spend in each plan in the year ahead.

"Our goal is that every state website will have the information to help you understand your real out-of-pocket costs," said Marc Boutin, president of the National Health Council, which offered its own calculating tool for customers during the last enrollment.

But with all the computer mishaps in the first enrollment year, neither the 36 federal nor 15 state exchanges had such a tool in the first year. Colorado tried in the first year, but consumers found the tool confusing and the exchange disabled it, said Adele Work, director of product implementation for Connect for Health Colorado. Consultants are working on a replacement, she said, but it may not be available in time for November. It's not clear which, if any, other states will have such a tool in place either.

Exchanges also did poorly in providing two other categories of information of great interest to consumers. Many exchange websites were unable to offer up-to-date lists of the medical providers who were in each network plan. And very few exchanges – Colorado and Nevada were exceptions – could tell consumers which medications each health plan covered, information that could make a difference of thousands of dollars.

Because of last year's disastrous roll-out, most exchanges will have modest ambitions for the second enrollment period. Offering consumers a smooth enrollment experience is the goal of most exchanges. But a smooth experience won't necessarily be enough to guarantee landing the best policy.


This piece originally appeared at
Stateline, an initiative of the Pew Charitable Trusts, where Michael Ollove is a staff writer. 

An ACA Subsidy Smoking Gun?

Robert VerBruggen - July 25, 2014

Up until this point -- see here and here -- I've said two things about the Obamacare text at issue in the subsidy cases: The formula quite clearly states that the subsidies are calculated based on the premiums charged in an "Exchange established by the State," but it's unclear whether this was an intentional restriction (designed to withhold subsidies from states that don't set up exchanges) or just a mistake. Under precedent, if the text doesn't make Congress's intent clear, courts can investigate the law's history to determine what Congress meant or defer to the bureaucracy's interpretation.

Well, now we have good evidence that the law's architects -- or, as Incidental Economist blogger Bill Gardner pointed out to me on Twitter this morning, at least some of the law's architects -- in fact intended to withhold subsidies. Via a commenter at the Volokh Conspiracy, the Competitive Enterprise Institute's Ryan Radia (who's contributed to RealClearPolicy) stumbled upon this 2012 video of an address by Jonathan Gruber, one of the law's driving forces:

At 31:25, Gruber says:

What's important to remember politically about this is if you're a state and you don't set up an exchange, that means your citizens don't get their tax credits-but your citizens still pay the taxes that support this bill. So you're essentially saying [to] your citizens you're going to pay all the taxes to help all the other states in the country. I hope that that's a blatant enough political reality that states will get their act together and realize there are billions of dollars at stake here in setting up these exchanges. But, you know, once again the politics can get ugly around this.

Again, this doesn't necessarily mean that all of the law's architects understood the text this way, and if different drafters thought the text meant different things, courts could simply defer to the administration's interpretation. The drafters who've come forward all swear that they meant the subsidies to be available everywhere -- and the law is so poorly written that a wide variety of readings are possible. For example, as discussed in the D.C. Circuit ruling and dissent, you could read the text to mean that federal exchanges can't have customers at all, because people in states without exchanges are not considered "qualified individuals" and the law quite arguably only authorizes exchanges to serve qualified individuals.

But at the very least, this statement ought to lay to rest the notion, popular among some on the left, that this setup is so far-fetched and ridiculous that Congress couldn't possibly have intended it. My favorite example here is from Ezra Klein, who used an analogy to federal highway funding to demonstrate how crazy it would be to create a program and then deny access to states that don't cooperate. The federal government hasn't followed his analogy exactly -- it hasn't built highways and then refused to let anyone drive on them -- but it has in fact, in a textbook example of "if you don't like the strings, don't take the money" federal policymaking, used highway funds to force compliance on state alcohol laws, just as some say Congress planned to use ACA subsidies to force states to create exchanges. 

This video might not make the ultimate outcome of these lawsuits easier to predict. But it certainly will make the debate in the months ahead a lot more interesting.

[Update: Peter Suderman found a TV appearance from this week in which Gruber claimed it was "unambiguous that (the wording of the law) was a typo." John Sexton found another 2012 speech in which Gruber also talked about subsidies being withheld. In an interview with Jonathan Cohn, Gruber explained:

I honestly don't remember why I said that. I was speaking off-the-cuff. It was just a mistake. People make mistakes. Congress made a mistake drafting the law and I made a mistake talking about it. ...

At this time, there was also substantial uncertainty about whether the federal backstop would be ready on time for 2014. I might have been thinking that if the federal backstop wasn't ready by 2014, and states hadn't set up their own exchange, there was a risk that citizens couldn't get the tax credits right away. ...

But there was never any intention to literally withhold money, to withhold tax credits, from the states that didn't take that step. That's clear in the intent of the law and if you talk to anybody who worked on the law. My subsequent statement was just a speak-o-you know, like a typo.

There are few people who worked as closely with Obama administration and Congress as I did, and at no point was it ever even implied that there'd be differential tax credits based on whether the states set up their own exchange. And that was the basis of all the modeling I did, and that was the basis of any sensible analysis of this law that's been done by any expert, left and right.

I didn't assume every state would set up its own exchanges but I assumed that subsidies would be available in every state. It was never contemplated by anybody who modeled or worked on this law that availability of subsides would be conditional of who ran the exchanges.

]

Robert VerBruggen is editor of RealClearPolicy. Twitter: RAVerBruggen

Immigration: Stopping the Surge

Michael Cipriano - July 25, 2014

As President Obama and Congress deliberate how to address the surge of unaccompanied minors at the nation’s southern border, David Inserra, a research associate at the Heritage Foundation, scolds the government for its lack of enforcement of immigration laws.

We took a few minutes to speak with Inserra. The conversation has been edited lightly for clarity.


What is the William Wilberforce Trafficking Victims Protection Reauthorization Act, and how has it contributed to the ongoing immigration surge?

The William Wilberforce Trafficking Victims Protection Reauthorization Act was a bill that was passed in 2008. It added a provision to the law that set up a different way of handling unaccompanied alien children from countries other than Mexico, non-contiguous countries. It meant that people who were from Central America or from the rest of South America, not including Mexico, had to be handled by the Department of Health and Human Services, not by the Department of Homeland Security.

The law contributed to making it more difficult to remove these individuals, because it provides that an alien child in the custody of HHS will placed in the "least restrictive setting" in the best interest of the child. About 90 percent of the time, that is the child's family, which can be located anywhere in the U.S. -- and this results in approximately 46 percent of unaccompanied alien children not showing up for their court date. They also have to go through the lengthy deportation process, which could take over a year.


What else is contributing to the surge? Don't we have laws that are supposed to deal with issues like these?

I would say there are two parts to that answer. First, there's the Deferred Action for Childhood Arrivals (DACA) memorandum and other uses of prosecutorial discretion, that is, President Obama's ability to say, "I am not going to enforce the law against certain groups." This is encouraging more individuals to come to the United States illegally. They don't necessarily understand the complexities of U.S. immigration law. They simply understand that people are being allowed to stay, certain illegal immigrants are not being deported, and that is acting as encouragement.

To the other part of your question -- whether we have ways of dealing with this -- I would certainly agree. In this paper, I note that there is a section of the Immigration and Nationality Act which does allow the president to remove any or all aliens that he comes across. It is in the sole, unreviewable discretion of the Department of Homeland Security. They simply don't want to use that authority. So even though there is this William Wilberforce Law that is clouding the issue, I believe he does have the authority to deal with this right now, but he simply does not want to do so.


Do you think Texas governor Rick Perry's decision to send the national guard to the border is a legitimate response to the crisis?

I do think that sending about 1,000 national guardsmen to support the Texas Department of Public Safety is an appropriate response as a short-term solution. Guardsmen are great at helping Border Patrol, providing logistics support, providing surveillance support. 

But they are not a good long-term fix. They are not as cost-effective as the Border Patrol, and they aren't fully trained to do all the things which a Border Patrol agent is trained to do. So right now, helping people get back to the front lines, that is helpful. But not as a long-term solution.


What long-term policies should be enacted to deal with the crisis. 

First of all, we should rescind DACA and other policies that the president has used to not enforce the law against certain people because they don't fit his "enforcement priorities." Enforcement priorities simply mean that some people aren't priorities and we aren't going to enforce the law against them. That is an additional magnet for people to come here illegally. Once they are in, they are pretty much good. 

Another thing we should be doing is to clarify the William Wilberforce law. That will help streamline the process and make it clear to the president that he does have the authority to remove these kids in an expedited matter. 

In the long term we need to talk about building partner capacity in Latin America. There is a push factor -- conditions encouraging immigrants to leave -- in Latin America, and we do need to make sure we are cooperating with those countries, providing security assistance down there to stop the crime and corruption that is going on.


Michael Cipriano is a RealClearPolitics editorial intern. 

Charles Ornstein, ProPublica - July 24, 2014

For months, journalists and politicians fixated on the number of people signing up for health insurance through the federal exchange created as part of the Affordable Care Act. It turned out that more than 5 million people signed up using Healthcare.gov by April 19, the end of the open-enrollment period.

But perhaps more surprising is that, according to federal data released Wednesday to ProPublica, there have been nearly 1 million transactions on the exchange since then. People are allowed to sign up and switch plans after certain life events, such as job changes, moves, the birth of a baby, marriages and divorces.

The volume of these transactions was a jolt even for those who have watched the rollout of the ACA most closely.

"That's higher than I would have expected," said Larry Levitt, senior vice president for special initiatives at the Kaiser Family Foundation. "There are a lot of people who qualify for special enrollment, but my assumption has been that few of them would actually sign up."

The impact of the new numbers isn't clear because the Obama administration has not released details of how many consumers failed to pay their premiums and thus were dropped by their health plans. All told, between the federal exchange and 14 state exchanges, more than 8 million people signed up for coverage. A big question is whether new members will offset attrition.

ProPublica requested data on the number of daily enrollment transactions on the federal exchange last year under the Freedom of Information Act because the Obama administration had declined to release this information, a key barometer of the exchange's performance, to the public. The administration also has not put out any data on the exchange's activity since the open enrollment period ended.

The data shows so-called "834" transactions, which insurance companies and the government use to enroll new members, change a member's enrollment status, or disenroll members. The data covers the 36 states using the federal exchange, which include Texas, Florida, Illinois, Georgia and Michigan.

When Healthcare.gov rolled out last fall, insurance companies complained that the information in the 834s was replete with errors, creating a crisis at the back-end of the system.

Between April 20 and July 15, the federal government reported sending 960,000 "834" transactions to insurance companies (each report can cover more than one person in the same family). That includes 153,940 for the rest of April, 317,964 in May, 338,017 in June and 150,728 in the first 15 days of July. The daily rate has been fairly stable over this period.

It was not immediately clear how many of the records involved plan changes or cancellations and how many were for new enrollments.

An insurance industry official estimated that less than half of the transactions are new enrollments. The rest are changes: When an existing member makes a change to his or her policy, two 834s are created — one terminating the old plan and one opening the new one.

Charles Gaba, who runs the website acasignups.net that tracks enrollment numbers, estimates that between 6,000 and 7,000 people have signed up for coverage each day on the federal exchange after the official enrollment period ended. Gaba's predictions were remarkably accurate during the open enrollment period.

"That doesn't account for attrition. That doesn't mean that they paid," Gaba said. "That's been based on limited data from a half dozen of the smaller exchanges, extrapolated out nationally."

The federal data obtained by ProPublica confirm some other facts about the rollout of Healthcare.gov, which was hobbled initially by technical problems. The slowest day was Oct. 18, when no 834 transactions were sent. That was followed by Oct. 1, the day the website launched, when a grand total of six records were sent to insurers.

By contrast, the busiest day was March 31, the official end of open enrollment, when 202,626 "834" reports were sent to insurers. The entire last week in March was busy.

About 86 percent of those who signed up for coverage on the federal exchange were eligible to receive government subsidies to help lower their monthly premiums. Those subsidies are being challenged by lawsuits in federal court contending they aren't allowed by the Affordable Care Act.

Two federal appeals courts came to conflicting decisions Tuesday on the permissibility of the subsidies (one said yes; the other no). They will remain in effect as the cases proceed in the courts, the Obama administration said.

The next time that the general public can sign up for coverage through the exchanges is from November 15 to February 15, 2015.

This piece originally appeared at ProPublica, where Charles Ornstein is a reporter, and was co-published with NPR's Shots blogClick here to download the data (Excel or CSV) released to ProPublica under the Freedom of Information Act. Read ProPublica's previous coverage of the Affordable Care Act and share your story.

Obamacare Suffers a Blow

Robert VerBruggen - July 22, 2014

The D.C. Circuit Court of Appeals has ruled 2-1 that the law's subsidies cannot legally be given out through the federal exchanges, as opposed to the state-run exchanges.

As I explained last week, Obamacare gives a formula for calculating premium subsidies, and that formula relies on the prices of plans available through "an Exchange established by the State under 1311 of the Patient Protection and Affordable Care Act." About three dozen states are using the federal exchanges instead of creating their own -- and it's section 1321, not 1311, that authorizes the federal government to step in when a state fails to set up an exchange. A literal reading of the law, therefore, suggests that subsidies are not legally available in much of the country.

However, as I also explained, precedent gives courts an "out" when an administration wants to interpret a statute broadly -- courts can decide that the law doesn't make Congress's intent sufficiently clear, and then turn to other clues to determine that intent. So this case partly hinges on the question of whether Congress wanted to deny subsidies in states that failed to set up exchanges. This is a much trickier issue -- it's plausible that Congress would have done this to create an incentive for the states to set up exchanges, but the law's architects deny that this was their intention and there's little direct evidence either way. (See the previous post for more details.)

At any rate, the D.C. appeals court reached the obvious conclusion regarding the text -- "a federal Exchange is not an 'Exchange established by the State'" -- despite finding that 1321 exchanges could otherwise be treated as 1311 exchanges. (This argument stems from the fact that section 1321 says the federal government may set up "such Exchange," referring to the exchange otherwise set up under section 1311. As the dissent notes, though, this is a hard case to make while maintaining that federal exchanges are not "established by the State," considering that section 1311 specifically defines an exchange as something established by a state.)

The court then responded to a variety of arguments that go beyond the text. For example, the court emphasized that legislative history must play a secondary role to the text itself, and noted that the evidence on this particular provision is scant. It also noted (as I did in my previous post) evidence that Congress at least considered the idea of intentionally limiting the availability of subsidies.

What happens next? Well, the government is sure to keep fighting, and another appeals court upheld the subsidies today, creating a circuit split. It seems likely that the Supreme Court will end up stepping in eventually (you can see the gory details of the process here). And over the past week or so we've run several pieces in our morning updates explaining what will happen policywise if federal subsidies are shut down.

Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen

Robert VerBruggen - July 21, 2014

Over at The New Republic, Alec MacGillis alleges that America's gun laws are helping to fuel the border crisis: Central American gangs buy guns here through straw purchasers, and then smuggle the weapons back home. He notes the stiff opposition the Bureau of Alcohol, Tobacco, Firearms, and Explosives encountered when it required gun stores in the four states along the Mexican border to report purchases of two or more long guns, as well as the political difficulty one would have trying to expand the rule nationwide.

He has a point. There is a gun-running problem, and the multiple-sale requirement is encountering more resistance than it probably should. But are guns from the U.S. really to blame, to any significant degree, for Central American violence? I have some issues with the way MacGillis goes about arguing they are:

According to data collected by the ATF, nearly half of the guns seized from criminals in El Salvador and submitted for tracing in the ATF's online system last year originated in the U.S., versus 38 and 24 percent in Honduras and Guatemala, respectively. Many of those guns were imported through legal channels, either to government or law enforcement agencies in the three countries or to firearms dealers there. But a not-insignificant number of the U.S.-sourced guns -- more than 20 percent in both Guatemala and Honduras -- were traced to retail sales in the U.S. That is, they were sold by U.S. gun dealers and then transported south, typically hidden in vehicles headed across Mexico, though sometimes also stowed in checked airline luggage, air cargo, or even boat shipments. (Similar ratios were found in traces the ATF conducted in 2009 of 6,000 seized guns stored in a Guatemalan military bunker -- 40 percent of the guns came from the United States, and slightly less than half of those were found to have been legally imported, leaving hundreds that were apparently trafficked.)

To MacGillis's credit, he does a much better job of using the ATF data than most journalists do -- he notes that these are not percentages of all crime guns recovered in each country, but merely percentages of guns seized and submitted to the U.S. for tracing. Unfortunately, though, this concession undermines the rest of his analysis. The ATF itself cautions that the data MacGillis uses do not constitute a "random sample"; guns are more likely to be submitted to the U.S. if foreign law-enforcement agencies have a reason to believe they actually came from the U.S.

Variations in the percentage of these guns that really did come from the U.S., therefore, don't mean much -- most likely, they just show that some countries are more thorough in sending trace requests and/or less skilled when it comes to guessing whether a gun is American. Neither do variations in the percentage of U.S.-sourced guns that came from retail stores -- this number will be affected not only by the flow of trafficked guns (the numerator) but also by the flow of legally traded guns (which makes up the rest of the denominator). The only number MacGillis provides that's useful is the one from the Guatemalan military bunker: It's a single collection of guns, so the stat may not be generalizable, but at least the denominator is a collection of illegal guns found naturally.

There are some other ways to analyze the ATF's trace data, however. First of all, we might want to get a better sense of the raw numbers. These confirm another contention MacGillis makes, which is that Mexico has a far bigger problem with American guns than Central America does (he cites both proximity and Mexico's more restrictive domestic gun laws). The biggest Central American source of gun traces that led to U.S. retail stores in 2013, Guatemala, submitted 133 of them. All six countries combined had only 391. By contrast, in most years, more than 5,000 guns from Mexico are traced to retail purchases from the U.S.

We can also try dividing the number of guns traced to U.S. retail stores by other statistics -- here, I'll try each country's total population and its total number of intentional homicides. These calculations assume, of course, that trace data reflect the overall prevalence of trafficked guns in a country. In fact, it's likely that some countries are more likely than others not to bother submitting guns that were in fact trafficked -- or that some countries have better law enforcement and find a higher percentage of all trafficked guns -- and this will throw off the calculations. So, while I think this is a useful exercise, take the results with a grain of salt.

By looking at total population, we can get a sense of which countries have the biggest appetite for illegal U.S. guns on a per capita basis. Belize seems to be an outlier thanks to its small size (20 gun traces but a population of less than a third of a million):

It's also helpful to see how each country's U.S. gun problem stacks up to its overall homicide problem. Note that this is not the percentage of homicides resulting from guns trafficked from the U.S. -- not all traced guns were used in homicides, so that number might be lower than what you see here, and not all homicide guns are found or traced, so it might be higher too. It's just a way of comparing the two problems:

So, every year, Central America as a whole traces one gun to a U.S. retail store for every 100,000 people living there and every 50 intentional homicides that occur. No question, that's much higher than the ideal number, which is zero. But is it high enough to make U.S. gun laws a significant driver of Central American violence, especially considering the far higher numbers for Mexico (about one for every 25,000 population and almost one for every five homicides)? It's hard to say: We don't know how accurate of a portrait these numbers paint, and even if we had better data it would be hard to say how effective stricter gun laws would be at staunching the flow of guns, not to mention how big of a homicide reduction a reduced flow of American guns would bring. [Update: ... and not to mention how many refugees would stay home thanks to a reduction in homicides. There's some evidence the border crisis isn't mainly driven by violence. Hat tip to Greg Pollowitz at National Review.]

You can download a spreadsheet containing the data for this post here.

Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen

Jerry Brown Provides a Powerful Argument for California Separation

Ian Adams, R Street - July 18, 2014

Gov. Jerry Brown's spokesman has raised questions about the feasibility of the Six States proposal, arguing that "the proposal has serious practical challenges." But does this position harmonize with Brown's own articulated preferences?

Venture capitalist Tim Draper believes his initiative to carve California into six separate states has now received the requisite number of signatures necessary to qualify for a November 2016 ballot. If it does, since the vote itself is not imminent, voters in California will have a substantial amount of time to consider separation and subsidiarity.

Public reaction to the initiative has been mostly critical. A poll in February placed opposition to the Six States plan at 59 percent. Among pundits and the Internet crowd, there is even less enthusiasm. In rare spasms of political conservatism, Ezra Klein of Vox believes that it is "an incredibly dumb idea"; the San Francisco Chronicle maintains that such a move will create a "state of confusion"; and Reddit's response has been the comically apoplectic: "F*** everything about this." Perhaps the difficulty Vladimir Putin faces in reassembling the former Soviet Union is informing their opinions.

Nonetheless, the obstacles to the Six States effort truly are daunting. First, the initiative would necessitate a majority vote of California's electorate. Second, approval of the split by the California Legislature, as it is currently composed, would likely be required. Third, the plan would need a green light from Congress. Each step appears a degree of magnitude more difficult to achieve than the last.

Still, the Six States plan deserves serious consideration if for no other reason than the existing precedent for such a conversation.

While California is today the most populous state in the nation, with 38 million people currently living within its borders, extreme geographical and cultural variations have always been a hallmark of the state. In 1859, California's legislature recognized the problem associated with such dissimilarities and passed legislation that would have broken California into two states, Northern and Southern. When the question was put to voters, they agreed. Three out of four voters cast a ballot in favor of devolution (Congress never acted on the question, preoccupied as it was by the events leading to the Civil War).

Today, there are any number of angles from which to evaluate the Six States proposal, including representational, economic and cultural.

The subsidiarity angle, though, is well-known as a favorite of California's current governor, Jerry Brown. Gov. Brown is a former Jesuit novice fond of Catholic social philosophy. During both the 2013 and 2014 State of the State addresses he referenced the notion of "subsidiarity." The idea behind subsidiarity is that matters should be handled by the least centralized and smallest competent authority possible.

Through the lens of subsidiarity, California's government is something of a failure. Can it really be said that 38 million Californians are represented in Sacramento by the smallest competent authority possible? Consider that a California state senator represents 931,000 people, compared to a California member of Congress who represents 704,000 people. This ratio of population-to-legislator is by far the highest in the United States.

Perhaps the issue could be addressed by something short of dividing the state since adding legislators is not unheard of, but other problems defy straightforward resolution.

Not unexpectedly, even though the U.S. Constitution grants the states plenary power over matters not explicitly reserved to the federal government, the State of California maintains state functions that are unable to respond flexibly to regional concerns. This is particularly true with regard to the state's regulatory environment on business, natural resources and transportation issues.

For example, those areas of California that are dependent upon cultivation of agricultural development, and those that are dependent upon the use of natural resources, chafe under the political weight of parts of the state capable of supporting themselves through other means. By use of the governor's subsidiarity concept and dividing California, new states would be able to establish regulatory environments reflective of not only their own political will, but also the economic assets of their region. Doing so, residents of the new states would have their will better reflected by a state government in closer proximity to their concerns – a total win for subsidiarity.

Clearly, the practical uncertainty associated with splitting up California demands at least as much attention as the theoretical justifications for a split. No doubt, much time and effort will be expended in that very endeavor. Still, an advocate of free markets and the plain benefits of federalism would be hard-pressed to ignore the virtues of reassessing California's present situation.

Ian Adams is an attorney in Sacramento, Calif. and an associate policy analyst of the R Street Institute. This piece originally appeared on the R Street blog.

Blog Archives