Cato Online Forum

Two Relatively Painless Ways to Boost Growth

By Robert E. Litan
November 2014

There is no one magic silver bullet that will permanently raise the potential growth rate of the U.S. economy — the driving force behind continued improvements in living standards — but I do offer two relatively painless ways of doing the job, and conceivably helping to make the result a bit more equitable to boot.

The first idea should be no stranger to Cato readers, to most economists, and to a bipartisan consensus at least in the U.S. Senate: greatly boost the numbers of permanent work visas (citizenship would be a plus) for high-skilled immigrants. Two classes of immigrants are especially important: foreign graduates with STEM (science, technology, engineering or math) degrees from U.S. universities, and immigrant entrepreneurs qualifying for a potential new expanded number of such visas.

The reason high-skilled immigrants are important for growth is that they are linked to the launch of new businesses, and it is startups that have been disproportionately responsible for disruptive innovation around which many other technologies are built (think the car, the airplane, computers, air conditioning, and Internet search). A substantial body of research has shown that not only are immigrants generally more likely to launch new business than native Americans, but that immigrants account for about a quarter of all tech startups, almost double immigrants’ share in the population (13 percent)1.

Dane Stangler and Jason Wiens of the Kauffman Foundation have calculated that the immigrant visa provisions in the Startup Act 3.0, which has not passed either congressional chamber, and has more liberal entry requirements than the Senate-passed comprehensive bill, would raise GDP by 1.5 percent over a decade, or roughly 0.15 percent a year (while creating 1.6 million new jobs). These estimates do not include the growth impact of enhanced STEM graduate visas.2

One widely discussed element of the comprehensive immigration reform package, an increase in the numbers of H1-B visas, would be good for tech employers who can’t find qualified American employees, but would not likely to be a major boost to growth. That’s because the H-1B is temporary (6 years, with a possible renewal), and thus offers no way for its recipients to launch their own businesses.

The second idea stems from the landmark decision by a California state court in June 2014, Vergara v. State of California, which held that the teacher unions’ tenure system violated the equal protection clause of the state’s constitution. The reasoning is that teacher quality is critical to student achievement and the court cited clear evidence that minority children were taught by less able teachers. Without the ability to remove those teachers, the local school system (and the state) were violating the children’s rights, in much the same way that “separate but equal” was held to violate the 14th amendment of the U.S. Constitution in Brown v. Board of Education of Topeka, Kansas.

Clearly, this decision — which is likely to be a roadmap for similar litigation and I believe similar outcomes in other states — has a lot to do with rectifying inequalities which affect students from their very earliest ages and then through the rest of their lives. But what does it have to do with growth?

A great deal, I submit. Students who are penalized by an educational system stacked against them virtually from the time they enter school until they graduate (or drop out, as all too many will), represent a huge waste of human capital. Let’s not forget that these are real people, with the same distribution of native talents as the rest of the population. But apply a poor educational environment with poor teachers (on top of a difficult socioeconomic and often dangerous physical environment), and the whole distribution shifts to the left: far fewer “stars” will make it out of this system to generate the next wave of innovations that could power growth.

Likewise, a huge range of students in the rest of the distribution could make a valuable contribution to society, but instead all too easily fall victim to a life of gangs or crime later in life, at huge cost to society (the costs of incarceration, more about this soon) and to their victims. Others suffer the economic and social loss of disappointed expectations. In economists’ speak, the nation suffers a huge opportunity cost.

The Vergara decision points the legal way toward a much better outcome, but the fiscal implications of the decision (and hopefully many like it in the future) have yet to be fully recognized. If school systems are going to get permission to pay teachers for performance that means that teaching eventually will no longer be the “safe” profession it has long been, with tenure protecting weak performers. Instead, teaching will be like private sector jobs where there is a risk of being let go for poor performance. With more risk attached to the job, it is likely that school systems will need to pay more in general.

But they are especially likely to have to pay premiums to attract good teachers into the crime-ridden areas where low income and minority kids often attend school, and to stay to teach there to teach in difficult circumstances — not just spend a two year Teach for America stint in such situations.

Where is the money going to come from? It need not come from more taxes, although I personally would not be opposed to that outcome as a last resort source of funds given the huge social importance of rectifying schooling inequalities. But my first resort would be to utilize savings from decriminalizing marijuana and embarking on early release programs for non-violent offenders now in jail.

There are now over 2 million people behind bars, of whom over 60 percent are serving time for nonviolent offenses. In particular, over a quarter of incarcerated Americans — more than 500,000 people — are there for drug offenses.3 Some 40,000 people are behind bars on marijuana charges, and approximately 650,000 arrests for marijuana-related offenses occur every year.4

The figures on cost per prisoner are all over the map, but a safe assumption for minimum security prisons is $25,000 per year, according to the Urban Institute.5 Freeing all inmates held on marijuana charges would alone save $1 billion, and a concerted effort to reduce incarceration rolls for nonviolent offenses could achieve savings of many billions more. Imagine what could be accomplished if we reallocated, say, $2 billion from incarcerating drug and other nonviolent offenders to paying teachers more. That money could finance an across-the-board annual pay raise of about $660 for all 3 million teachers,6 or we could use the same amount and pay an additional $10,000 a year for 200,000 teachers or $20,000 for 100,000 of them.

Put something like these tradeoffs up for a vote, and I’ll bet you’d have plenty of teachers siding with a lot of other voters — dare I say a majority in many, perhaps most states. Add the high-skill immigration proposal I started with and we’d be well on our way to appreciably lifting our nation’s rate of potential economic growth.

1 For a guide to the literature, see
2 For a summary of the study and links to the full study, see
3 See
4 See
5 See
6 See

The opinions expressed here are solely those of the author and do not necessarily reflect the views of the Cato Institute. This essay was prepared as part of a special Cato online forum on reviving economic growth.

Robert Litan is a nonresident senior fellow at the Brookings Institution. He is the author of Trillion Dollar Economists (2014), among many other books and articles on economics and public policy.