Archives: 05/2019

Four Missing Components from Kushner’s Merit-Based Immigration Reform

President Trump recently released another plan to reform the immigration system.  The latest plan was drafted with the help of his son-in-law and presidential advisor Jared Kushner.  Overall, this last plan is the best of those proposed so far as it reduces legal immigration the least, as my colleague David Bier ably explains.  Kushner’s impact on President Trump’s thinking on immigration, as on the area of criminal justice reform, has been extremely positive and I hope that his advising role continues to expand.  The fundamental idea of merit-based immigration reform is sound: There should be more economic immigrants admitted to the United States.  Still, there are at least four additions to the Kushner plan that would make an enormously positive difference in the final policy. 

The following four points are reforms that should have been included in the Kushner plan that would transform the U.S. immigration system into one that emphasizes economic and merit-based immigration without reducing family-based immigration.  Regardless of any other legal changes, the following four reform suggestions would be positive on their own.



  1. Reform employment-based green cards by exempting immediate relatives, exempting adjustments of status, and ending the per-country limits from the numerical quota.  Exempting children and spouses from the employment-based green card quotas would immediately increase the number of green cards for skilled workers by over 60,000 annually, based on the 2017 numbers.  Even better, Congress should exempt adjustments of status from the green card quota as 82 percent of them were already living here legally on another visa before earning their employment-based green cards.  At a minimum, this would increase the number of employment-based green cards issued annually by over 100,000 and possibly by far more.  Lastly, ending the per-country limits would reduce unjust discrimination against immigrants from populous countries.  While ending the per-country limits would not increase merit-based immigration on its own, exempting family-members and adjustments of status would achieve a similar benefit for those in the backlog while boosting the number of employment-based green cards annually.
  2. The Kushner plan should have endorsed the state-based visa reform program that was introduced last Congress by Senator Ron Johnson (R-WI).  Senator Johnson’s bill was inspired by the Canadian Provincial Nominee Program (PNP) that is the second largest source of economic immigrants to Canada.  Johnson’s state-based visa program would issue 500,000 non-immigrant visas annually divided among the states by population with each state guaranteed at least 5,000 annually.  Unlike other visas, state-based visas would be mostly regulated by state governments that would determine the economic criteria for selecting migrants.  Rather than be constrained by the visa categories that the federal government believes are the best, states would be allowed to experiment by creating visas for entrepreneurs, investors, or workers of any skill or type.  Crucially, states would not have to participate in the state-based visa program if they don’t want to and the number of visas would automatically grow in states that follow the rules and shrink in those that don’t.  Please check out Senator Johnson’s bill and David Bier’s wonderful paper about it.
  3. The third proposal that would have boosted the merit of Kushner’s plan is the immigration tariff.  Heavily influenced by the late Nobel Prize-winning economist Gary Becker, an immigration tariff would sell non-immigrant permanent resident visas called a gold card.  The numerical quantity of visas would not be limited in law, but the high prices for each visa would act as a de facto numerical limit on the number issued.  As I proposed in a mock tariff schedule, Congress could set the price for visas based on the age and education of the migrant whereby younger and better-educated migrants pay a lower price than older less educated migrants.  The price of the visa offers some labor market protection to American workers but is also flexible enough that more migrants will purchase the gold card when the U.S. economy is expanding and fewer will do so when the economy is shrinking.  Furthermore, those willing to pay high prices for a visa will be extra motivated to succeed here and they will also be more likely to believe that they have the abilities to do so.  The money raised through this tariff, which I estimate here, would be more than enough to cover the cost of the surge of asylum seekers along the southern border and provide additional resources for border security – although I prefer to use the funds to finance a tax cut for lower-skilled Americans.  There are many other proposals in this vein, such as the IDEAL visa, that the administration should also consider to boost merit-based migration.
  4. Do not cut the number of immigrants.  The Kushner plan does cut the number of new immigrants who will arrive in the United States annually even though it keeps the number of green cards the same.  The Kushner plan transfers family-sponsored preference green cards and diversity green cards to the employment-based green card categories.  Ninety-four percent of the immigrants who received a family-sponsored preference green card and 98 percent of those who received a diversity green card were new arrivals while the rest adjusted their statuses from other visas.  Meanwhile, eighty-two percent of employment-based green cards went to migrants who were already legally working and living in the United States – mostly on H-1B visas.  In other words, the Kushner plan trades new immigrant arrivals in the family and diversity green card categories for green cards in the employment-based green card categories that will go to workers who are mostly already in the United States.  Don’t get me wrong – green cards are better than H-1B visas as workers are more productive, have more flexibility, and make longer-term investments in their new home when on green cards as opposed to the temporary H-1B visa, but trading new immigrants for improving the economic prospects of existing migrants is a bad deal.  Although the number of green cards remains the same under the Kushner plan, the number of new immigrants arriving in the United States will decrease across all visa categories.




Many meritorious immigrants to the United States do not enter on economic visas but come in on family-based green cards and the diversity visa.  In the same way that Americans can be family members and workers, many immigrants enter the United States on family-based green cards also intend to work.  Over time, they are also becoming more skilled.  The share of admitted immigrants who have at least a college education increased from 22 to 39 percent from 1993 to 2015.  Over the same period, the share of admitted immigrants who are high school dropouts decreased from 37 percent to 27 percent.  Since so few new immigrants to the United States are on employment-based green cards, much of the increase in education must come from family-based immigrants and those on the diversity visa.  

Setting the number of legal immigrants is not like setting a budget that needs to be balanced over the long term.  There is no deficit here.  Increasing the number of explicit economic or merit-based immigrants does not mean that the number of family-based green cards needs to be cut.  It’s wonderful that the president seems to have moved away from his previous position that the number of green cards needs to be reduced and we can thank Mr. Kushner’s hard work for that, but the government can increase the number of economic and merit-based immigrants without changing any other portion of the immigration system.  If President Trump wants to be known for shifting America toward a more merit-based immigration system, boosting the number of economic immigrants without cutting family immigration is the path of least resistance and greatest gain.

Most Canadian Immigrants Are Not Economic Immigrants, they are Family-Based Immigrants

President Trump recently released another plan to reform the immigration system.  It was crafted with the help of his son-in-law and presidential advisor Jared Kushner.  My colleague David Bier ably explains the details of the Kushner plan, but I want to focus on a common misconception that lays at the justification for merit-based immigration reform: In reality, there is not a single country in the OECD or European Union (EU) where economic immigrants are a majority of new permanent residents, even in merit-based systems like Canada’s. 

The one-pager for the Kushner plan says that 63 percent of Canadian permanent resident visa are granted in employment and skill-based categories compared to just 12 percent in the United States, 57 percent in New Zealand, 68 percent in Australia, and 52 percent in Japan.  Those percentages all include the derivative family members of the immigrant workers admitted through the employment and skill-based visa categories.  In other words, economic or skills-based immigrants are allowed to bring their immediate relatives but they also count against the economic visas even though they are not the economic immigrants themselves.  The Kushner plan one-pager counts the derivative family-members of merit-based economic immigrants in Canada, New Zealand, Australia, Japan, and the United States as skilled immigrants.  This grossly exaggerates the percentage of immigrants who are economic or skills-based immigrants and undercounts those who are family-based immigrants.

Figure 1 shows the percentage breakdowns of immigrants granted permanent residency whereby the family-members of employment or skills-based economic immigrants are counted against the All Family visa category.  With the exception of Japan, every country mentioned in the Kushner one-pager grants more family-based immigrants permanent residency than to explicitly economic or skills-based immigrants.  These data comes from the OECD and are pooled for the years 2005-2016.  Table 1 breaks down these figures for all OECD and EU countries and shows that there are only three countries where there is more family-based immigration than economic immigration:  Japan, Spain, and Ireland.


Figure 1: Immigrants by Broad Visa Category


Table 1: Immigrants by Broad Visa Category


This problem is known in the United States.  In 2017, a minority of employment-based green cards actually went to workers while the majority went to their minor children and spouses.  Other countries or the OECD tend to count the family-members of economic skills-based immigrants in the same way – inflating the percentage who appear to be skilled economic immigrants and deflating those who are family members.  Just to be clear, derivative family members often work too and are an economic asset, but they are the types of immigrants whom the proponents of merit-based immigrants want to cut.

There are many points in favor and against creating a more merit-based immigration system in the United States, but let’s not be confused by the admittedly odd way that governments or the OECD count economic and family-based immigrants.  At a minimum, let’s stick to the facts when comparing immigration systems across countries:  No country in the OECD or EU admits mostly economic immigrants. 

New Report Exposes Jones Act’s Flaws

Evidence of the Jones Act’s failures continues to mount. Just weeks after the release of an OECD study predicting substantial gains from the law’s repeal, the Congressional Research Service (CRS) has a new report out which places the law’s shortcomings in sharp relief.

The report’s description of U.S. shipbuilding is particularly eyebrow-raising. Rather than bolstering this sector through the Jones Act’s domestic build mandate, the CRS notes the sector has experienced a steady deterioration in competitiveness since the law’s passage:

A 1922 government report on shipbuilding indicated that U.S.-built ships cost 20% more than those built in foreign yards. The cost differential increased to 50% in the 1930s. In the 1950s, U.S. shipyard prices were double those of foreign yards, and by the 1990s, they were three times the price of foreign yards. Today, the price of a U.S.-built tanker is estimated to be about four times the global price of a similar vessel, while a U.S.-built container ship may cost five times the global price, according to one maritime consulting firm.

Unsurprisingly, this loss of competitiveness has translated into commercial ship output that amounts to barely a rounding error in total global production:

The Merchant Marine Act of 1970 (P.L. 91-469) added as an additional objective of U.S. maritime policy to have a merchant marine “supplemented by efficient facilities for building and repairing vessels.” U.S. shipyards typically build only two or three oceangoing ships per year, and none for export, so they do not achieve economies of scale. There may be gaps of several years in between orders for container ships. In recent years, the demand has been sufficient to sustain one shipyard that builds only commercial ships. However, this yard stated that its employment had fallen below 100 people and that it had no vessels under construction or on order as of March 31, 2019.

The relationship between the Jones Act and the commercial shipbuilding sector’s travails is almost certainly causal. Drawn to a captive domestic market, U.S. shipyards have not sought to compete internationally and thus cannot achieve scale. This focus on the U.S. market means fewer opportunities for specialization and related productivity gains, while the lack of international competition reduces the need for improvement and innovation.

The result is spiraling costs and decreased demand for U.S.-built ships, leaving shipyards with less output to spread their fixed costs across. It’s a vicious cycle that shows no sign of ending. 

These spiraling shipbuilding costs aren’t just an economic problem. While the Jones Act is often justified on national security grounds, the high cost of domestic shipbuilding encouraged by this law has also impaired the military’s ability to source its sealift ships from U.S. shipyards: 

The cost differential is also an issue for Department of Defense officials in charge of military sealift ships. As discussed later in this report, the military has modified a plan to build sealift ships domestically, finding it unaffordable, and instead will buy more used foreign-built cargo ships. Since U.S. shipyards do not build vessels for export, they are not required to compete with foreign shipyards on price or vessel characteristics.

Indeed, just last week the Secretary of the Navy said that he “can’t really afford a lot of $400 million ships when I can go out and buy used [roll-on/roll-off ships] for $35 to $40 million.” And if high ship costs are a deterrent to notorious spendthrift Uncle Sam, the effect is surely no less pronounced within the broader maritime industry.

These high ship acquisition costs, as well as operating costs at least 2.7 times those of foreign-flagged vessels, help explain why demand for U.S. coastwise shipping has declined despite the many advantages of ocean transport:

While domestic ships are carrying fewer tons of freight today than they did in the 1950s, their most direct competitors, railroads and pipelines, are carrying more. Domestic ships have lost market share to land modes even though ships have economic advantages. Ocean carriers do not need to acquire and maintain rights-of-way like railroads and pipelines. They can move much more cargo per trip and per gallon of fuel than trucks and railroads. Although ships are slower than truck and rail modes, many shippers are willing to sacrifice transit time for substantially lower costs, as long as delivery schedules are reliable.

Also seemingly explained is the lack of coastal shipping to transport containers arriving aboard large cargo ships from abroad. Instead of being placed on smaller ships as part of a hub-and-spoke system, they are placed on rail and roads, with the latter leading to increased highway congestion:

Transshipment of international containerized cargo by feeder ships is prevalent abroad, but the practice does not exist in the United States. The Jones Act would require such ships be U.S.-built, -crewed, and -owned. Lack of transshipment services increases demand for rail and road connections to ports, as smaller feeder container ships do not play a role in distributing international containerized cargo among U.S. ports.

Not everyone, however, can take refuge from the Jones Act through the use of overland forms of transportation, and the CRS report notes that the Jones Act fleet tends to “operate in markets where shippers have little alternative.” This means the non-contiguous states and territories of Alaska, Hawaii, and Puerto Rico which have little choice but to suffer the law’s exorbitant costs. 

Meanwhile, high ship acquisition costs also undoubtedly play a role in the absence of entire ship types from the Jones Act fleet, and the advanced age and insufficient quantities of those that do exist:

Not all ship designs are represented in the Jones Act fleet. “Project cargo” or “heavy-lift” vessels are often used to carry oversized pieces of equipment such as smaller vessels, ship engines and modules, wind turbine parts, and power generation equipment. They would be useful for moving dredging fleets to project sites. There have not been any such vessels in the Jones Act fleet in recent decades. The Department of Defense has used “national defense” waivers of the Jones Act to move radar systems and newly built vessels on foreign-flag heavy-lift vessels. This type of cargo typically does not generate regular shipments in any one region; thus these ships would likely need to extend their market reach beyond the United States to include the international market. However, the higher cost structure of Jones Act operators is an obstacle to competing for international shipments.

Two dry bulk ships are in the oceangoing Jones Act fleet, and they appear to be mostly inactive, possibly because they are nearly 40 years old. This is twice the economic life of a ship in the global fleet (where ships are typically sent for scrapping between 15 and 20 years of age). The sole Jones Act-qualified chemical tanker was built in 1968. No LNG tankers are in the Jones Act fleet despite new domestic markets as a result of the shale gas boom. The lack of sufficient Jones Act-qualified tanker capacity to move booming shale oil production coastwise added to pressure for lifting the crude oil export ban in 2015.

One result is that despite the economy’s continued expansion, and thus increased transportation needs, the number of ships and amount of cargo they carry are in long-term decline:



The report is rife with examples of the Jones Act’s inability to meet its stated aims. It points out, for example, that the law “has not succeeded in meeting the stated policy goal of sustaining a growing merchant marine that carries an increasing proportion of the nation’s commerce.” The objective of “providing shipping service on all routes essential for maintaining the flow [of commerce] at all times” and having the “safest” types of vessels are similarly unmet, with the CRS stating that “the Jones Act fleet does not appear to achieve either of these goals.” The report adds that “One can also question whether the policy objective of having ‘the best equipped and most suitable types of vessels’ has been achieved.”

The failure of protectionist policy is one of the world’s more predictable phenomena, and prior to its passage some appeared to foresee its problems and urge a different course. As the CRS report points out, the minority report to a 1919 House committee report to the bill that would become the Jones Act called for an approach based on competition and the removal of restrictions:

…in order to build up and sustain an American merchant marine it is absolutely necessary to remove every restriction against American merchants acquiring ships, whether built in the United States or out of the United States, at the lowest possible price, in order to enable them to compete with other nations in the transportation of the commerce of the world…Our American iron and steel manufacturers were unable to compete until they had to. When they had to they did compete successfully. Our shipbuilders can and will do likewise.

And more than 50 years ago the Lyndon B. Johnson administration accurately stated that protectionism and federal largesse would not reverse the U.S. merchant marine’s dwindling fortunes:

At a 1967 congressional hearing, Alan Boyd, Secretary of Transportation in the Lyndon B. Johnson Administration, testified that the U.S. merchant marine was “too small, too old, and too unproductive,” and stated, “you do not revitalize an industry by flooding it with Federal dollars and imprisoning it within a wall of protection.”

These voices, however, have been consistently ignored in favor of those advocating yet more protectionism and government intervention, the very policies that led to the U.S. maritime sector’s current predicament. But their record of failure has never been starker or more plain to see. It’s time for a new approach based on the principles of free markets, openness, and competition whose record of success is unsurpassed. It’s a plan just crazy enough to work.  


What Factors Should an Immigration Points System Include?

Last week, President Trump backed a plan that would create a new legal immigration category based on “points.” The idea is borrowed from immigration systems in several countries, including Canada, which award points to applicants based on various personal characteristics (language skills, educational attainment, family ties, etc.). The Canadian government, for example, establishes the cap on visas for the year, and applicants with the highest point total receive one of the cap slots for that year.

Advantages of a points system

Congress should not cap skilled legal immigration, which provides massive economic and fiscal benefits to the United States. Assuming it will anyway, however, a points system could be a fair and economically beneficial way to allocate some green cards, but policymakers should avoid some of the common pitfalls in implementing one.

Ideally, a points system should serve a single purpose: economic growth. Trying to enlist a points system for other goals—family reunification, humanitarianism, “assimilation,” etc.—ultimately results in an incoherent system that doesn’t serve any of them well. To be clear, this is not to say that these other goals should receive no representation in the immigration system anywhere, just not in a points system.

First, the main reason a points system makes sense (in the context of caps on skilled immigration) is that the points can consider characteristics that will result in greater long-term economic growth. A pure employer sponsorship model only considers the benefits to a single employer in the short term, but a points system can more adequately account for broader economic effects.

Second, a points system has the virtue of not requiring an employer to file anything with the government. Untying employers from the regulatory bureaucracy would be a massive improvement over the current system. Third, a points system stops long queues and waiting periods for employers from developing by issuing green cards to whoever had the most points in that application period, quickly connecting workers and employers. Fourth, if demand exceeds supply, a points system fairly and intelligently awards green cards without resorting to a lottery.

Factors a points system should consider

1. Job offers: This criterion should be almost mandatory. In order to increase economic growth, a person must work. Of course, Congress could adopt an expansive definition of “work” to include self-employed entrepreneurs and investors, but it should not get into the business of selecting immigrants for businesses. Employers should lead the hiring process. One mistake that architects of a points system can make is focusing too heavily on immigrants’ resumés rather than whether the U.S. market actually demands their services.

In 2015, Canada reformed its points system to weight job offers much more heavily after discovering that many immigrants with high qualifications on paper (at least according to the government) failed to find employment. In 2013, Canadian immigrants with university degrees earned 33 percent less than their Canada-born counterparts. In the United States, foreign-born college graduates earn more than U.S.-born. Any points system should be led by the market, not bureaucrats.

2. Earnings: A points system should prioritize applicants with the highest wage offers (or expected income from U.S. investments or businesses). Higher wages generally indicate higher productivity. Higher productivity means more economic growth. Under the U.S. system that limits legal immigration from particular countries (the “per-country limits”), higher-paid immigrants from India and China actually end up waiting longer than other immigrants. A points system prioritizing higher wages would also incentivize employers to offer the highest wage that they can to their immigrant applicants, guaranteeing that immigrants receive the market wage. Unfortunately, Canada does not rely on wages to prioritize job offers.

3. Ages: As with the job offer, it should be almost mandatory that applicants in a points system be under the age of 60, even if they have a high wage offer. Average immigrant wages in the United States almost double from their early 20s to their late 40s and decline precipitously thereafter. Retirees are on average very fiscally negative for the U.S. government. The decline actually occurs earlier among the highest earners (Figure 1).

This means that a points system should prefer a younger worker with a somewhat lower wage offer to an older worker with a somewhat higher wage offer—both because the younger worker’s productivity will rise while the older worker’s will fall and because the younger worker simply has many more productive years left. Again, that’s not to say that Congress couldn’t design a retiree visa. It just shouldn’t use a points system for that purpose. 

Figure 1: Average Wage and Salary Income by Age For Immigrants by Education

4. U.S. work experience: While immigrants with job offers indicate that a U.S. company demands their services at one point in time, a points system should give extra consideration to immigrants with a history of employment in the United States. First of all, U.S. employment experience (usually on a temporary visa of some kind) proves that the immigrant lived up to expectations in the labor market.

Second, a period of employment proves that the immigrant wants to remain in the United States. Immigrants who secure a green card under the points system but then discover that the United States is not to their liking undermine the entire purpose of the system. U.S. immigrants have a fairly high emigration rate—which is partially due to faulty government policies, but also a result of immigrants choosing to leave voluntarily. For both reasons, prior U.S. employment demonstrates that the worker will likely continue to contribute to economic growth.

5. Minor children: According to the National Academy of Sciences (NAS) 2017 report, the second generation (e.g. children of immigrants) is the most fiscally positive generation. A points system would be wise to award more points to immigrants who have children who are automatically be eligible to immigrate with the primary applicant. A U.S. birth rate below replacement level is a critical threat to economic growth, as Europe is discovering, so a points system concerned with long-term growth should credit immigrants who bring children or have children here.

Many children of U.S. immigrants will contribute greatly to economic growth in a few years. In fact, 83 percent “of the finalists of the 2016 Intel Science Talent Search, the leading science competition for U.S. high school students, were the children of immigrants.” Again, because they have their entire lives ahead of them, children of immigrants provide another source of long-term economic growth.

6. Educated spouses: Under any immigration process, immigrants will always be permitted to bring their spouses with them, so any points system will need to evaluate the spouses of the applicants. Spouses shouldn’t count against the points system cap, but it should consider how an immigrant’s spouse could contribute to economic growth.

Evaluations of spouses should start with the same factors as the primary applicant, but because they are often following their spouses to the United States, spouses may not have jobs lined up in advance, so instead of focusing solely on wages and job offers, the system should award more points for immigrants with highly educated spouses. Immigrants with college degrees are much more likely to find employment in the United States than those who haven’t graduated college, and college graduation predicts higher wages among immigrants.

Factors a points system should ignore

Most proposed points system incorporate a variety of irrelevant factors. These include:

  • English language proficiency
  • Educational attainment
  • Occupation
  • International acclaim or prestigious awards
  • Family ties
  • Civics knowledge
  • Place of birth, religion, race, etc.

While educational attainment, English language proficiency, and certain high demand occupations are associated with higher productivity, they are not nearly as good of a measure as the actual wage offered to the worker. These factors could be useful in evaluating the likely productivity for someone without a job offer yet (as in the case of the spouses), but a good points system should maintain a job as a baseline for everyone (but the spouse and minor children) for that reason. Civics exams, country of origin, and family ties are totally irrelevant to determining someone’s productivity. A points system should never consider these factors at all.

The Trump points proposal

We don’t have the full details of Trump’s proposal, but the president’s speech last week provided a few lines about the White House plan. He said it would favor younger workers with an offer of employment, prioritizing those with higher wages. That’s a great start, but unfortunately, his plan also includes several irrelevant factors including preferences for people with higher education, a government-determined “valuable skill,” English language proficiency, and U.S. civics knowledge.

The vague White House slides about the plan describe three categories eligible for the points system—immigrants with “extraordinary talent,” those in “professional and specialized vocations,” and “exceptional students.” A points system would not need to separate categories in this manner unless the government plans to include irrelevant factors like which universities an applicant graduated from or which fields that they work in.

While there is no reason it should come at the expense of other immigration channels, a points system could be useful addition to the U.S. immigration system. The government should avoid the temptation to select the immigrants with the best resumés and make sure that the points are driven primarily by the private sector. The Trump plan appears to incorporate some of the important elements of an effective points system, but also includes elements that will undermine it. Hopefully, Congress can help the White House iron out these issues as it evolves it into actual legislation.

Wherefore the Freedom Caucus?

In a column for Reason Magazine yesterday, Matt Welch asks “What’s the point of a ‘limited government’ bloc that doesn’t limit government?” Indeed, in the Trump era some of the President’s most strident defenders can be found amongst the ranks of the Freedom Caucus, and, as my colleague Chris Edwards points out, they seem every bit as comfortable with big deficits as the other fiscal-conservatives-cum-spendthrifts in the GOP.

But, to my knowledge, nobody has yet performed a systematic analysis of the Freedom Caucus’ voting behavior vis-a-vis other Republicans in the House. Do they, as a caucus, even vote cohesively? If so, are they at all differentiable from generic Republican House members? I set out to test this using the NOMINATE methodology to assign an “ideal-point” estimate for each member of the House during the modern era of Republican dominance (2011-2018). For a quick explainer on NOMINATE, see this page. The basic upshot is that it will differentiate members of Congress according to their voting patterns, along two dimensions, with the x-dimension being dimension of primary importance and capturing inter-party variation. Democrats tend to have negative scores on the x-dimension, and Republicans tend to have positive scores. The y-dimension captures intra-party variation. The sign is arbitrary, but the inter-point distance is meaningful. I’ve labeled everyone who appears on the “membership” list on the Freedom Caucus wikipedia page. 

First, let’s look at all final passage votes over this period, regardless of substantive topic (this excludes votes such as amendments and procedural votes).

passage.png (862×550)

Freedom Caucus members are clearly differentiable from generic Republicans, and predominantly occupy the upper-right quadrant. 

Now, let’s generate ideal point estimates over subsets of all substantive rollcalls (amendments, procedure, etc.) pertaining to certain select topics:

domestic_appropriations.png (862×550)entitlements2.png (862×550)

environment.png (862×550)defense.png (862×550)

Across each of these rollcall subsets, Freedom Caucus members are differentiable from other House Republicans, and consistently occupy the upper-right quadrant (note: comparing the substantive implications of these two dimensions across subsets is difficult. Nonetheless, Freedom Caucus members are distinct in some sense).

 It’s worth noting that on procedural votes, House leadership exercises more effective discipline on its members. This is consistent with the political science literature (Cox and Poole 2002). For example, look at the pattern on motions to recommit:

 recommits.png (862×550)

While the substantive implications of the above graphs aren’t clear as of yet, I don’t think we can so easily dismiss the Freedom Caucus as generic Republicans. 


Philadelphia Soda Tax

A recent New York Times article reports that today’s Democratic primary in Philadelphia partially represents a referendum on the city’s soda tax, even though it is not on the ballot. Incumbent Mayor Jim Kenney, the engineer of the 1.5 cents per ounce tax on sugary and artificially sweetened beverages which took effect in 2017, is facing two Democratic challengers who both support repealing the controversial tax.

Since its inception, the tax has faced vigorous opposition from a coalition of the beverage industry, grocery store owners, and Teamsters, as well as citizens and politicians who worried that it would unfairly burden the poor. Supporters have touted the tax’s revenue-raising potential and the public health benefits of taxing soda in a city with some of the highest obesity and Type 2 diabetes rates in the United States.

Kenney has always presented the tax as a revenue-raising measure and he reports that it has raised $200 million for education and infrastructure programs. While Kenney’s stated goal was never to change soda drinking behavior, many supporters of the tax have focused on the potential health benefits and studies have found mixed results of its impact on soda consumption in the city. From an economic standpoint there are, among others, three serious concerns about the tax’s ability to both raise revenue and reduce obesity.

First, as I discussed in 2016 shortly after it was passed, both the level and scope of the tax mean it has a dubious impact on reducing obesity. The tax is on soda consumption in general which burdens a majority of soda drinkers who are not obese instead of directly incentivizing people to avoid all behaviors that lead to excessive weight (a better alternative would be to price health insurance for the obese higher, but this is not allowed under the Affordable Care Act). And the 1.5 cents per ounce rate is estimated to decrease body mass index by .26 points, a tiny amount when considering that obesity is defined as a body mass index of 30. The tax is inefficient and it’s doubtful it has had any significant impact on obesity.

Second, it is likely that many soda drinkers have responded to the increased soda prices, but not necessarily by reducing their soda consumption. A paper by Mabel Andalon and John Gibson, which I reviewed in the winter 2017/2018 issue of Regulation, looked at Mexico’s nationwide soda tax and found that consumers substituted for cheaper brands of soda. Between 2012 and 2014 average soda prices increased by 11.9 percent, but the average price of purchased soda increased by half that rate, suggesting consumers simply bought lower-priced soda. This response is likely present in Philadelphia as well, where consumers are just as able to substitute to lower-price or even untaxed alternatives (in fact, the Times article reports that there has been a spike in the sales of powdered drink mixes which are not taxed). This undermines both the health benefits and (in the case of untaxed substitutes) the revenue raised.

Finally, people can simply avoid the tax by buying their sugary drinks in nearby localities without their own soda taxes. In the current issue of Regulation I reviewed another paper, by Stephan Seiler, Anna Tuchman, and Song Yao, that observed this exact taxpayer response in Philadelphia. The authors found that beverage purchases in Philadelphia decreased by 42 percent, but this reduction was fully offset by increased purchases outside of Philadelphia.

So, while the positive impacts of the tax are doubtful, the negative effects are felt most by Philadelphia stores and the poorest citizens who were already buying the cheapest sodas or don’t have the ability to buy their beverages outside of the city.

However, I mentioned in my 2016 post that there are some positives of the tax:

The program was not sold to voters as a public health measure, but rather as a means of raising new tax monies. The discussion of the tax and the public spending for which the revenues would be used was explicit.  And the tax is a consumption tax rather than a tax on the rich or corporations.  To be sure, the tax is on a very narrow consumption base and thus is distortionary, but at least the tax is visible. The voters will see the tax and the public services that result and can make an informed decision in the next election about the tax and its uses.

Of course, there are other issues up for debate in this primary and it is unlikely that the soda tax alone will make or break Mayor Kenney. But the tax’s transparency has allowed for an effective democratic process and voters will be able to weigh the soda-tax-funded programs relative to its incidence.

Written with research assistance from David Kemp.

Chinese Ships on the Mississippi River: Just Another Jones Act Tall Tale

Did you know that the Jones Act prevents Chinese ships from sailing on the Mississippi River? That, at least, is what Rep. Brian Babin (R-TX) claimed in a recent speech on the House floor. For dramatic effect the congressman used a picture of a ship flying an oversized Chinese flag with St. Louis’s Gateway Arch prominently displayed in the background:


“This is a hypothetical picture, thank goodness,” said the Texas congressman. “A Chinese-built vessel, subsidized by their communist regime, operated by the Chinese and delivering Chinese goods all in the very heartland of the United States of America. But this could easily become a reality if the Jones Act is waived.” 

Other Jones Act supporters have made similar warnings. Clay Maitland, the chairman of the Merchant Marine Policy Coalition, stated last week that his support of the Jones Act was “all about preventing the Chinese government from getting control of our inland waterways,” while in 2017 the president and CEO of Jones Act carrier Tote Inc. said that the law’s repeal could result in North Korean barges and tugboats operating on the Mississippi River. And just last year the American Maritime Partnership, a Jones Act advocacy group, ran the following internet ad:


Leaving aside the desirability of foreign ships operating on the Mississippi River, does the Jones Act actually prevent this from happening? Well, no. The Jones Act restricts the domestic waterborne transport of goods to vessels that are U.S.-flagged, U.S.-built and at least 75 percent U.S.-crewed and owned. It says absolutely nothing, however, about the ability of foreign ships to operate on inland waterways. 

In fact, there are foreign ships operating on U.S. inland waterways at this very moment. Indeed, a quick peek reveals numerous such vessels on the Mississippi River as far north as Baton Rouge and still more on the Delaware and Columbia Rivers headed to and from Philadelphia and Portland. 

That foreign ships are not more prevalent on inland waterways is a matter of simple physics: they’re too large. The ship used in Rep. Babin’s hypothetical picture, the CSCL Africa, has a draft of 11 meters (36 feet). The Mississippi River around St. Louis lacks the depth to support such a vessel. This helps explain why the U.S. Army Corps of Engineers only maintains a 9-foot shipping channel north of Baton Rouge but a 45-foot shipping channel south of it. 

But even the idea of smaller-sized foreign vessels on the country’s inland waterways is largely a figment of the Jones Act lobby’s imagination. A 1999 U.S. International Trade Commission report which examined the Jones Act’s economic impact said that it did not bother including inland waterways in its analysis because U.S. operators there “do not appear to be significantly vulnerable to foreign competition that may occur in the absence of Jones Act restrictions.”

This is in large part because foreign vessels would still have to comply with U.S. laws and regulations unrelated to the Jones Act. As Michael Hansen of the Hawaii Shippers Council points out:

…[T]he inland waterways barge industry is the least threatened by Jones Act reform or outright repeal. Even a full national repeal of maritime cabotage would be an insufficient condition for achieving continuous employment of foreign flag vessels operating in purely U.S. domestic trade. This is especially true for foreign shipping to displace purely domestic inland waterways traffic. This is primarily due to the extra-cabotage legal and regulatory web enveloping the inland waterways – from immigration, customs, taxation, business registration, labor, health and safety, to wage and hour – which would prevent a foreign flag vessel from being continuously employed in domestic service.

In addition, U.S. builders of smaller vessels such as barges are much more internationally competitive. This means that inland waterways operators are able to purchase vessels at prices similar to those found overseas, thus boosting their ability to compete. As the USITC states, “…operators in the inland trade acquire vessels from the internationally competitive U.S. barge and smaller shipbuilders and so have substantially more competitive capital costs.”

Jones Act or not, Chinese ships or North Korean barges cruising the American heartland’s innermost waterways isn’t going to happen. The American people deserve more facts and less scaremongering.