Stock

Eric Gomez and Benjamin Giltner

This is the second of a three‐​part series on the US arms sales backlog for Taiwan. The first article outlined our methodology for determining the capabilities in the backlog and showed how traditional military capabilities, which are more flexible but also more vulnerable to destruction, dominate the backlog.

This article examines another type of US military support for Taiwan that is distinct from backlogged weapons: maintenance support for US weapons already delivered to Taiwan.

We did not include maintenance in the arms sales backlog dataset because Taiwan already possesses the weapons that maintenance sales support. Unlike backlogged weapons, Taiwan can use the weapons supported by maintenance sales to defend itself. Moreover, the Stockholm International Peace Research Institute’s (SIPRI) Arms Transfer Database, which we used to determine the items in the arms sales backlog, does not include data on maintenance sales.

To create our dataset of maintenance sales we used Defense Security Cooperation Agency (DSCA) announcements of major arms sales. We added all maintenance sales announced between January 20, 2017 and November 1, 2023 to the dataset to compare the Trump and Biden administrations. Due to the difficulty of determining the timeline for final delivery of the spare parts and services included in maintenance sales, we did not eliminate any of the announced sales from our dataset.

We classified maintenance sales as asymmetric or traditional based on the type of weapon they support. As shown in Figure 1, between January 20, 2017 and November 1, 2023 the DSCA announced almost $3.5 billion of maintenance sales to Taiwan, with $923 million supporting asymmetric capabilities, and $2.5 billion supporting traditional capabilities.

Five of the twelve maintenance sales supported missile defense capabilities, with a total amounting to $1.8 billion or 54 percent of all US maintenance sales to Taiwan. If the missile defense system in question was immobile, such as Taiwan’s PAVE PAWS early warning radar, it is coded as traditional. If the missile defense system was mobile, such as a Patriot battery, it is coded as asymmetric.

On a dollar basis, the United States provided Taiwan with significantly more maintenance support for traditional capabilities than asymmetric capabilities. This makes sense, as traditional capabilities tend to have higher maintenance costs because they are used more frequently in peacetime than most asymmetric capabilities. Taiwan’s manned fighter aircraft, for example, have flown more missions to monitor increased Chinese military activity in Taiwan’s air defense identification zone.

Figure 2 shows the differences between maintenance sales under the Biden and Trump administrations. Both administrations spent much more on maintaining traditional than asymmetric capabilities.

The Trump administration sold slightly more maintenance support to Taiwan than the Biden administration, although the Biden administration has a little over a year left in its current term. As shown in Table 3, Biden has announced eight maintenance sales to Trump’s four. If these trends continue, we expect Biden’s total maintenance support sales to exceed Trump’s by the end of Biden’s term.

There are two concerning implications drawn from the data on US maintenance support for Taiwan.

First, the high costs of traditional capabilities does not stop at their acquisition. While traditional platforms perform a wider range of missions, this comes with more wear and tear and higher maintenance costs. If Taiwan’s defense budget was much higher, it could afford to pursue both asymmetric and traditional capabilities without a significant tradeoff. However, given Taiwan’s current level of defense spending, pursuing even small numbers of traditional capabilities comes with a significant opportunity cost in money not available to build and maintain asymmetric systems, which tend to have lower up front and lifetime costs.

Second, the amount of maintenance support the United States provides Taiwan suggests that Taiwan is not able to sustain its high‐​end capabilities on its own. This is a significant vulnerability in the event of a conflict, where the United States risks becoming limited or stopped from resupplying Taiwan. To address this vulnerability, the United States and Taiwan should explore options for boosting Taiwan’s ability to conduct maintenance of US‐​provided weapons with spare parts produced entirely in Taiwan. Additionally, both the United States and Taiwan should plan for stockpiling spare parts with a similar level of urgency as plans for stockpiling munitions.

Taiwan Arms Sales Backlog Excel File

Johan Norberg

In my essay Globalization: A Race to the Bottom – or to the Top?, part of Cato’s Defending Globalization project, I show that the belief that an open world economy would hurt working conditions turned out to be false. On the contrary, trade and investment are associated with higher wages, less child labor, safer working conditions, and better environmental performance. Between 2000 and 2016, the global rate of deaths related to work declined by 14.2 percent.

One of my examples is the effect of Bangladesh’s garment export industry in reducing poverty and creating better jobs, especially for women.

An interesting new study by Laura Boudreau of Columbia Business School, recently featured on the excellent podcast Trade Talks, sheds light on one transmission link between globalization and progress: how businesses and consumers in the West put pressure on suppliers to improve working conditions.

After the collapse of the Rana Plaza building, which killed at least 1,260 textile workers in 2013, foreign multinational apparel buyers decided to oblige every factory they bought from to introduce an Occupational Safety and Health Committee. Boudreau conducted a year‐​long field experiment with 84 supplier factories, randomly enforcing the mandate on half. She found that the commitment was not just PR. Because factories did not want to miss out on business, they formed committees, and these also correlated with small improvements in workers’ health and safety in those factories.

The conclusion is the complete opposite of what critics of capitalism have warned about for decades, when claiming that Western multinational companies (MNCs) take advantage of countries with weak governance to put downward pressure on standards. On the contrary, they often step in when governments fail. Boudreau writes: “In weaker states … it may be MNCs or multi‐​stakeholder coalitions with MNCs’ participation that provide enforcement. This research suggests that in such contexts, MNCs can contribute to increasing compliance with labor standards.”

For more on globalization’s race to the top and related issues, be sure to check out my full essay and all the other content on Cato’s Defending Globalization project page.

Adam N. Michel

New research shows that the 2017 tax cuts significantly raised business investment and boosted economic growth. This most recent study is the latest entry in a long catalog of economic research that finds taxes matter for investment, economic growth, and labor market outcomes. Despite this history of robust results, a recent American Compass report argues that the 2017 Tax Cuts and Jobs Act did not significantly impact the economy. More sophisticated economic analyses almost universally find the opposite: taxes matter for investment and growth.

Economists have known this for decades. In a 2012 academic literature review, Will McBride summarizes 26 studies on the empirical relationship between taxes and economic growth. He finds that “all but three of those studies, and every study in the last fifteen years, find a negative effect of taxes on growth.” The research also consistently finds that corporate income taxes are the most economically harmful.

In a more recent 2019 retrospective review of new empirical fiscal research following the 2008 financial crisis, Valerie Ramey shows that, almost universally, tax increases are estimated to reduce GDP. The most consistent result from the time series estimates reviewed shows that GDP declines by between two and three times the revenue the new taxes raise.

The most recent entry in this long list of research uses variations in how the 2017 tax reform impacted different corporations differently. The researchers estimate that the tax cut “caused domestic investment of firms with the mean tax change to increase by roughly 20% relative to firms experiencing no tax change.” The authors extrapolated their short‐​run estimates to assess the long‐​run effects of the law. Their results are consistent with some of the most optimistic modeling of the TCJA’s economic and revenue effects, finding significant increases in capital stock, productivity, and wages.

Numerous individual studies looking across multiple policy changes and using different empirical techniques come to similar conclusions:

Using a narrative method to determine fiscal policy changes, Christina and David Romer found that a tax increase equal to one percent of GDP has a “consistently negative” effect on the path of real GDP, and “the maximum effect is a fall in output of 3.08 percent after ten quarters.” They find the same tax increase results in an 11 percent decline in gross private domestic investment almost three years after the change. 
Using a structural vector auto‐​regression approach, Andrew Mountford and Harald Uhlig build on earlier research to find “a maximal present value multiplier of five dollars of total additional GDP per each dollar of the total cut in government revenue five years after the shock.” 
Combining empirical methods, Karel Mertens and Morten Ravn found that a one percent cut in the personal income tax rate boosts real GDP per capita by up to 1.8 percent. That translates to a multiplier similar to those described above of about 2.5. They found that a one percentage point cut in the corporate tax rate boosts GDP by as much as 0.6 percent and has “no significant impact on revenues.”
When investigating more targeted reforms to reduce effective tax rates on new business investments, Eric Zawick and James Mahon found that expensing “raised investment in eligible capital relative to ineligible capital by 10.4 percent between 2001 and 2004 and 16.9 percent between 2008 and 2010.” They also found that smaller firms and those with tight cash flow show the biggest investment response.

Ultimately, the economic history of any specific reform will be confounded by numerous concurrent events. In the case of the Tax Cuts and Jobs Act, its positive economic effects were undermined by President Trump’s trade policies, and COVID-19 truncated the economic record. Despite these headwinds, the reform’s key changes to the corporate tax rate and allowance for full investment deductions had a measurable positive effect, in line with the historical record.

Across this body of research, specific methods may be subject to critique, and the estimates are not all directly comparable; however, studies consistently show a positive effect of tax cuts on investment and economic growth. Economic theory also supports these results. When you lower effective tax rates on work and investment, you should expect to get more of each.

Kayla Susalla

It is frequently asserted that public schools educate all students. However, students deemed at‐​risk are sometimes transferred to alternative schools, also known as Alternative Education Campuses (AECs). AECs are understudied, yet existing data pose concerns for the outcomes and practices of the students they serve.

In the 2021–22 school year, over half‐​a‐​million students were enrolled in ACEs. ACEs are public schools commonly serving students deemed at risk of educational failure. Habitual suspensions, risk of expulsion, poor grades, substance abuse, truancy, pregnancy, or other challenging circumstances may result in a referral from a traditional public school, though students can also transfer voluntarily. Referral to an AEC is at the discretion of the traditional public school, and assignment can be temporary or permanent.

A survey conducted by the National Center for Education Statistics (NCES) found a majority of teachers in AECs are hired directly to teach in such schools, but teachers can also transfer voluntarily and involuntarily from traditional public schools. AECs can focus on behavioral issues, credit recovery, or offer specific pedagogical approaches, though others may serve as a broader alternative for students in traditional public schools. Charter schools, as well as private schools contracted by school districts, can serve as AECs.

AECs can play a pivotal role in the trajectory of a child’s education. Individualized learning plans, lower teacher‐​student ratios, and flexibility in format are a few of the assets AECs can possess. However, there is concern that traditional public schools refer underperforming or difficult students to AECs to evade low graduation rates, and disproportionately transfer minority students.

A Government Accountability Office (GAO) report highlights the disparities in demographics and disciplinary actions of students who attend AECs. The GAO found in the 2015–2016 school year that Hispanic boys, black boys and girls, and boys with disabilities had higher enrollment shares in disciplinary alternative schools than in non‐​alternative schools. Additionally, the report noted, “in every district GAO visited, officials said students had experienced multiple types of trauma, such as gang violence, death of schoolmates or parents, poverty, or homelessness.”

In comparing disciplinary measures in alternative schools between the 2013–14 and 2015–16 school years, the report found disabled students experienced a 17 percent increase in occurrences of corporal punishment. Black students experienced an increase in corporal punishment, as well as in‐​school suspensions, school‐​related arrests, and referrals to law enforcement, while White and Hispanic students had a decrease in all types of discipline.

It is concerning if traditional public schools are referring at‐​risk students to schools designed to address behavioral or academic problems, and if in those schools harsh disciplinary measures like suspensions, law enforcement interventions, and corporal punishment are in widespread use. Such policies push kids, who are already struggling in and outside of school, further behind academically and can potentially re‐​traumatize them.

Preceding that, discretionary referrals may be influenced by conscious or unconscious bias, and may contribute to disproportionate referrals for specific groups of students.

Research evaluating outcomes of students who attend an AEC is scarce, but what exists is not encouraging. As seen in the graphs below, graduation rates tend to be extremely low compared to traditional public schools, as well as proficiency in math and language arts.

These results notwithstanding, some research suggests AECs produce generally positive outcomes. But it is unclear if such outcomes are due to lowered standards for AEC students, or effective practices.

A meta‐​analysis found a small positive effect on school performance, attitude, and self‐​esteem, suggesting students enjoy attending AECs. But it found no effect on juvenile delinquency. Wilkerson et al. discovered students in academic‐​remediation alternative schools earned more credits versus students attending a traditional public school that fit the alternative school attendance criteria. This despite having lower attendance rates.

Wilkerson et al. noted some academic‐​remediation alternative high schools allowed students to graduate with fewer credits compared to the traditional public schools they would have attended. Lowering requirements pushes kids through the system, possibly leaving them unprepared after graduation, and defeating the purpose of education.

The study also found alternative schools were associated with increased suspensions, but fewer office discipline referrals, suggesting alternative schools are quicker to use exclusionary discipline.

Clear guidelines on what justifies a referral to an alternative school are vital to mitigate subjective procedures and potentially discriminatory practices. Additionally, if AECs are using harsh disciplinary measures and leaving kids unprepared, students should not be forced into schools potentially causing more harm. Ultimately, policies should allow funding to follow students to freely selected institutions fitting their unique needs.

Romina Boccia

With a 2023 deficit of $2 trillion and the Treasury needing to roll over about one‐​third of publicly held debt at now much higher interest rates within one year, legislators are beginning to more seriously grapple with the fiscal and economic costs of unsustainable government debt. Among the most promising options: empowering a fiscal commission to overcome the political gridlock plaguing Congress.

Despite a favorable retelling of history, modeling such a commission after the 1981 Greenspan Commission is bound to fail.

During a congressional hearing held by the House Budget Committee, on the need for a fiscal commission to help Congress solve the growing debt issue, former Senator Rob Portman (R‑OH) said,

“On the question of what’s worked and what hasn’t worked. BRAC has worked. The Base Realignment and Closure process. […] Another model would be the ’81 Greenspan Commission, which was a bipartisan group of experts and members of Congress.”

I’ve previously written about why BRAC, the successful Base Realignment and Closure commission model, offers the most promising path to avoiding a severe fiscal crisis. See here and here.

The Greenspan Commission, however, failed. Its members were unable to come to agreement within the commission structure, and the final product fell short of the goals the president had set for it. Using the Greenspan Commission as a model to develop recommendations to stabilize the debt is unlikely to work. Had it not been for the imminent pressure of indiscriminate cuts to Social Security bringing Congress to the table, we would likely remember the Greenspan Commission’s track record very differently.

A Brief History of the Greenspan Commission

President Ronald Reagan established the National Commission on Social Security Reform, as the Greenspan Commission was officially called, by executive order on December 16, 1981. Alan Greenspan ended up chairing the commission, which is how it got its name. Its goal was to avoid indiscriminate Social Security benefit cuts, scheduled to occur about a year later. As President Reagan instructed in the executive order, the commissioners were to:

review relevant analyses of the current and long‐​term financial condition of the Social Security trust funds;
identify problems that may threaten the long‐​term solvency of such funds;
analyze potential solutions to such problems that will both assure the financial integrity of the Social Security System and the provision of appropriate benefits;
and provide appropriate recommendations to the Secretary of Health and Human Services, the President, and the Congress.

The commission had 15 members, with five members each selected by the president and the House and Senate majority leaders (in consultation with their respective minority leaders). No more than three of the five commissioners chosen by each designated leader could be affiliated with their own party.

The Greenspan Commission was made up of seven members of Congress and eight outside experts, including some former members of Congress and presidential advisors. Perhaps its most compelling feature was that current members of Congress made up a minority of its members, with the majority consisting of former politicians, presidential advisers, and other non‐​elected experts. Despite this composition, the commission failed to generate solutions to resolve Social Security’s financing crisis as commission members could not agree on reforms. Instead, they were bailed out by a subgroup that formed at the last minute and that worked directly with leaders in the White House to negotiate a deal the commission ended up voting for.

The Greenspan Commission Failed

While the Greenspan Commission’s final report became the basis for H.R. 1900, the Social Security Amendments of 1983 (P.L. 98–21), how the commission arrived at this result, and the fact that Congress had to step in with amendments to resolve about one‐​third of the projected long‐​range actuarial deficit, illustrates the Greenspan Commission’s failures.

As reported by the nonpartisan Congressional Research Service:

“By the last formal meeting on December 10, 1982, the commission had failed to develop consensus recommendations. The impasse was broken when a smaller group of commission members and White House officials began to meet privately in December 1982. In the end, the recommendations in the commission’s final report were projected to resolve two‐​thirds of Social Security’s projected long‐​range funding shortfall, leaving Congress to resolve the rest.”

Robert M. Ball, former commissioner of Social Security from 1962 to 1973 who served as House Speaker Tip O’Neill’s representative on the Greenspan Commission, published a book about his experience on the Commission in 2011. In The Greenspan Commission: What Really Happened, Ball reported,

“Nothing […] should obscure the fact that the National Commission on Social Security Reform was not an example of a successful commission. The commission itself stalled […] after reaching agreement on the size of the problem that needed to be addressed. As a commission, that was as far as it got. […]

“To suggest that the Greenspan Commission provides a model for resolving questions … would be laughable if it were not so dangerous….”

There was a very particular fiscal context that ensured that certain members of Congress and White House officials came up with a backup plan when the Greenspan Commission failed. That context would not exist for any fiscal commission that today’s Congress might establish in the next year or two. America’s fiscal crisis has no certain trigger date and while there is broad agreement that the US is facing an unsustainable debt trajectory, its impact is subject to debate.

Social Security was within a couple of months, and perhaps within only a few weeks, from running out of legal borrowing authority as its trust fund ledger was about to run into the red. US government officials were up against a hard deadline with economically harmful and politically devastating consequences: Social Security benefits would be cut automatically once the trust fund reached zero. It was only in this context—to avoid a much scarier default scenario with a certain trigger date—that Congress and the White House came to an agreement and were able to push the 1983 Social Security amendments through the legislative process. They even followed regular order including public hearings and the ability to debate and amend the legislation in the committees of jurisdiction in both the House and the Senate.

An Approach That’s Bound to Fail (Again)

Were Congress to model a fiscal commission using the Greenspan Commission as a model, this approach would most likely fail, just like the original Greenspan Commission did. Relying on current politicians and political appointees, who are in the throes of political incentives, to produce an effective solution to reforming entitlement programs is a fool’s errand. And in today’s legislative context, there would be no hard, and harmful deadline to force an alternative agreement through the political process—the way that Social Security’s imminent trust fund exhaustion did in 1983. That’s why BRAC’s model, relying on independent experts and using a fast‐​track procedure that allows for silent approval in Congress (no affirmative vote necessary, no amendment, only the option to reject the proposal in its entirety), is much more promising because it provides more cover for Congress and takes politics out of the commission process.

Procrastination Comes With a High Cost

If Congress fails to act in the next couple of years, legislators will eventually face such hard, and harmful deadlines, when the Medicare and Social Security trust funds become exhausted, between 2031 and 2033. Unfortunately, American workers and taxpayers will pay a high price if Congress once again waits until the very last minute to avert automatic benefit cuts to old‐​age programs. By that time, benefit changes, including slowing the excess growth in future benefits, which take time to phase in, won’t be able to avert the imminent crisis. Only immediate tax increases and additional deficit financing will work, then and there.

This is why most Democrats are in no hurry to reform Medicare and Social Security. If your worldview is that Americans aren’t paying enough taxes and that federal government spending on old age benefit programs should be even higher, then waiting puts you in the best position to achieve those goals.

Instead of procrastinating to the detriment of economic growth and American living standards, Congress should empower a BRAC‐​like fiscal commission now. Acting sooner rather than later puts more options on the table and allows for gradual changes to benefit programs, without forcing economically damaging tax increases or further borrowing on taxpayers.

Jennifer Huddleston

So far, this month has seen the Biden administration issue a significant executive order and the US-UK Summit on artificial intelligence (AI) governance, as well as a growing number of bills in Congress. Much of the conversation around AI policy has been based on a presumption that this technology is inherently dangerous and in need of government intervention and regulation. Most notably, this is a major shift away from the more permission‐​less approach that has allowed innovation and entrepreneurship in other technologies to flourish in the US, and towards the more regulatory and precautionary approach that has often stifled innovation in Europe.

So, what would a soft touch approach look like? I suggest that to embrace the same light touch approach that allowed the internet to flourish, policymakers should consider four key principles.

Principle 1: A Thorough Analysis of Existing Applicable Regulations with Consideration of Both Regulation and Deregulation

Underlying much of the conversation around potential AI is a presumption that AI will require new regulation. This is also a one‐​sided view of regulation as it does not consider how existing regulation may also get in the way of beneficial AI applications.

AI is a technology with many different applications. Like all technologies, it is ultimately a tool that can be used by individuals with a variety of intentions and purposes. For this reason, many of the anticipated harms, such as the potential use of AI for fraud or other malicious purposes, may be addressed by existing law that can be used to go after the bad actors using the technology for such purposes, and not the technology itself.

Some agencies, including the Consumer Financial Protection Bureau (CFPB), have already expressed such views. It is likely that many key concerns, including discrimination and fraud, are already addressed by existing law and such interpretations should be focused on the malicious actor, which in most cases will not be the technology itself.

After gaining a thorough understanding of what is already covered by regulation, the appropriate policymakers could identify what — if any — significant harms are unaddressed or where there is a need for clarity.

In accounting for the potential regulations that may impact AI, policymakers should consider not only those areas but look at both harms that remain unaddressed by current law, as well as the need for deregulation in some cases where existing regulations may get in the way. This may include opportunities for sandboxes in highly regulated areas such as healthcare and financial services where there are potentially beneficial applications but existing regulatory burdens make compliance impossible.

Principle 2: Prevent a Patchwork, Preemption of State and Local Laws

As with many areas of tech policy, in the absence of a federal framework, some state and local policymakers are instead choosing to pass their own laws. The 2023 legislative session saw at least 25 states consider legislation pertaining to AI. The nature of this legislation varied greatly, from working groups or studies to more formal potential regulatory regimes. While in some cases states may be able to act as laboratories of democracy, in other cases, such actions could create a patchwork that prevents the development of this important information.

With this in mind, a federal framework for artificial intelligence should consider including preemption at least in any areas addressed by the framework or otherwise necessary to prevent a disruptive patchwork.

While regulatory burdens of an AI patchwork could create problems, some areas of law are particularly reserved for the states, where it would be inappropriate to have federal control. In some cases, such as the use of AI in their own governments, state and local authorities may be better suited to those decisions and even able to show the potential benefits of embracing a technology. With this in mind, preemption should be tailored to preserve the traditional state role but prevent the type of burdens that would not allow the same technology to operate across the nation.

One example of what such a model of preemption may look like could come from the way states have sometimes preempted city actions to ban home‐​sharing services like AirBnB while still allowing them their traditional regulatory role. Home‐​sharing and other sharing economy services have often been a point of friction in certain cities. But when large cities in a state ban these services or engage in overregulation, it can have a disruptive effect on the ability of individuals across the state to use or offer these services and can undermine good state policy on allowing such forms of entrepreneurship.

This approach prevents cities from banning or creating regulations, such as a de facto ban on home‐​sharing, but allows them to preserve their traditional role when it comes to concerns within their control, such as trash and noise ordinances.

A similar approach could be applied to potential preemption in the AI space, clarifying that states may not broadly regulate or ban the use of AI but may retain their ability to consider how and if they want to choose to use AI in their government entities or other traditional roles. Still, this should be done within the traditional scope of their role and not as a de facto regulation, such as banning the ability to contract with private sector entities if they are using AI or requiring actions that would result in being unable to have any product comply.

Principle 3: Education Over Regulation: Improved AI and Media Literacy

Many concerns arise about the potential for AI to manipulate consumers through misinformation and deep fakes, but these concerns are not new. We have seen in the past with technologies like photography and video that society develops norms and adapts to the way we find out the truth when faced with such concerns. Rather than try to dictate norms before they can organically evolve or stifle innovative applications with a top‐​down approach, policymakers should embrace education over regulation as a way to empower consumers to make their own informed decisions and better understand new technologies.

Literacy curriculums evolve with technological developments, from the creation of the internet to social media platforms. Improved AI literacy — as well as media literacy more generally — could empower consumers to be more comfortable with the use of technology and make sound choices.

Industry, academia, and government actions towards increased AI literacy have provided students and adults alike with opportunities to increase their confidence around and in artificial intelligence tools. Initiatives like AI4K12, a joint project of the AAAI and CSTA in collaboration with the NSF, are already working to develop national guidelines for AI education in schools and an expansive community of innovators to develop a curriculum for a K‑12 audience.

In higher education, many universities have started to offer courses on ChatGPT, prompt engineering, basic literacy, and AI governance. Research projects like the Responsible AI for Social Empowerment and Education (RAISE) at MIT work with government and industry partners to deploy educational resources to students and adult learners on how to engage with AI responsibly and successfully. In a society that is increasing its reliance on technology, innovators in every sector are providing a multitude of avenues to familiarize users of all ages with innovations, like artificial intelligence.

Many states are utilizing this approach as they propose and pass social media literacy bills. While many states have a basic digital literacy requirement, a few states are putting forth proposals with a focus on social media literacy. Florida passed legislation earlier this year mandating a social media curriculum in grades six through twelve. Last month, California’s Governor Newsom signed AB 873, a bill that included social media as a part of required digital literacy teachings for all students K‑12.

Other states such as Missouri, Indiana, and New York are considering similar bills that would require education in social media usage. Such an approach could be expanded to include AI as well as social media. Policymakers should also consider ways this approach could be applied to reach adult populations—through public service campaigns or simple opportunities like providing links to more information from a range of existing civil society groups or through public‐​private partnerships.

Principle 4: Consider the Government’s Position Towards AI and Protection of Civil Liberties

The government can impact the debate around artificial intelligence by restraining itself when appropriate to protect civil liberties, while also embracing a positive view of the technology in its own rhetoric and use.

Notably, during the Trump administration, an executive order was issued about the government’s use of AI that was designed to foster guardrails around key issues like privacy and civil liberties. It also encouraged agencies to embrace the technology in appropriate ways that could modernize and improve their responses. Further, it encouraged America’s development of this technology in the global marketplace.

However, the Biden administration seems to have shifted towards a much more precautionary view, focusing on the harms rather than the benefits of this technology.

Ideally, policymakers should look to see if there are appropriate barriers to using AI in government that can be removed to improve the services it provides to constituents. At the same time, much as with concerns over data, they should provide clear guardrails around its use in particular scenarios, such as law enforcement, and respond to concerns about the government’s access to data by considering data privacy not only in the context of AI but more generally.

Conclusion

Policymakers in the US decision during the 1990s to provide a presumption that the internet would be free unless regulation was proven necessary provided a great example of how a light touch approach can empower entrepreneurs, innovators, and users to experience a range of technological benefits that could not have been fully predicted. This light touch approach was critical to the United States flourishing while Europe and others stifled innovation and entrepreneurship with bureaucratic red tape.

As we encounter this new technological revolution with artificial intelligence, policymakers should not rush to actions based solely on their fears but instead look to the benefits of the approach of the previous era and consider how we might continue it with this new technology.

Colleen Hroncich

As a homeschooler, you don’t generally expect balanced coverage of homeschooling from mainstream media outlets. So my expectations for a recent Washington Post report on homeschooling were low—the bar was on the floor, as my daughter likes to say. I was pleasantly surprised to find that it was actually pretty evenhanded.

The report, “Home schooling’s rise from fringe to fastest‐​growing form of education,” is the latest installment in the Post’s Home‐​School Nation series. It’s likely the most in‐​depth analysis of homeschooling to date, and the big takeaway is that homeschooling is the fastest growing education model in the US. Homeschooling spiked during COVID-19 and then dipped a little, but the report estimates the number of children homeschooling was 45 percent higher in the 2022–23 school year than it was in 2017–18.

For the report, the Post collected homeschooling data from 2017–18 through 2022–23 for thousands of school districts across the country, conducted more than 100 interviews, and fielded a national poll of homeschool parents. Due to varied data collection practices in different states, the reporters were only able to compile statistics from thirty‐​two states and the District of Columbia. There’s a handy infographic that allows you to look up homeschooling stats by school district.

There are diverse homeschooling families and communities profiled in the report. Hillsborough County, FL, with more than 10,000 homeschoolers last year, could be considered the capital of American homeschooling, according to the Post:

“Today, Hillsborough home‐​schoolers inhabit a scholastic and extracurricular ecosystem that is in many ways indistinguishable from that of a public or private school. Home‐​schooled kids play competitive sports. They put on full‐​scale productions of “Mary Poppins” and “Les Miserables.” They have high school graduation ceremonies, as well as a prom and homecoming dance.”

While Florida’s 154,000 homeschoolers marked the largest count among states with data available, New York had the fastest growth. The homeschool population in New York has more than doubled since 2017, although its 52,000 homeschooled students are just one‐​third of Florida’s tally. Much of this growth came from New York City, especially Brooklyn and the Bronx, where some districts saw a more than 300 percent increase in homeschoolers. But the actual number of homeschoolers in the city is still small compared to district enrollment. On the flip side, the story looked at a district in rural Kentucky where more than 11 percent of students are homeschooled.

Although the piece included very strong examples of homeschoolers, the overall tone was one of skepticism. Consider this statement, for example, “Many of America’s new home‐​schooled children have entered a world where no government official will ever check on what, or how well, they are being taught.” This is rather absurd considering how poorly many students fare in schools that are run by government officials, not just checked on by them.

In Baltimore, where public schools spend more than $21,000 per student, reading and math scores are atrocious. Earlier this year, it was reported that thirteen Baltimore City high schools had zero students who tested proficient on the state math exam. Zero. In Detroit, only five percent of eighth graders tested at “proficient” or better in reading this year. Clearly, having a “government official” “check on what, or how well, they are being taught” doesn’t offer any guarantees.

While those are particularly egregious examples, the overall performance of public schools in the National Assessment of Educational Progress (NAEP)—often called the Nation’s Report Card—isn’t too rosy. This year, 64 percent of fourth graders were below proficient in math, as were 73 percent of eighth graders. Reading results were similar, with 66 percent of fourth graders below proficient, along with 69 percent of eighth graders. Again, these unimpressive results show oversight by government officials does not ensure student learning.

Then there’s the notion interwoven throughout the piece that our current public school system is the way things have always been and should always be. “[Homeschooling] could also undermine the role that public schools have traditionally played in American life.” But this model—grouping kids by age instead of ability and assigning them to schools based on where they live—is a relatively new experiment. For most of human history, education looked much more like homeschooling and microschooling.

And no piece on education alternatives would be complete without the implication that children belong to the public school system. Even though homeschooled families in Kentucky receive no taxpayer assistance, a Kentucky teacher who is head of her local union “worries about what home schooling’s growth will mean for the children in the public education system. ‘If [home‐​schooled] students are not enrolled in our district, we are not getting funding for them,’ she said. ‘And we are already underfunded.’”

But why would the district receive funding for kids who aren’t attending the schools? The report doesn’t delve into that question.

Homeschooling often seems like a radical approach when people first encounter it—even to people who eventually become homeschoolers (myself included). But in many ways, the current public school system is the radical approach. There’s no other area of society where we assign people to a service based on how old they are and where they live—even other levels of education, like preschool or college. There’s no other time in our lives where we’re so strictly segregated based on age. And there’s no other facet of life where government mandates funding, provision, and attendance of a good or service through a monopoly provider with the implicit assumption that one size fits all.

While homeschooling is still a small segment of education in the US, the explosive growth detailed in the Post’s report shows people are increasingly recognizing the benefits of more personalized education. I think that bodes well for our future.

Colin Grabow

For decades the federal government has been manipulating the domestic sugar market through a system of import and production restrictions that increases the price of sugar. This program is great for sugar producers, but—as a report released last week by the Government Accountability Office makes clear—it hurts pretty much everyone else. Once the sugar program’s costs are subtracted from its benefits, the report notes that various studies suggest a net loss to the economy ranging from $780 million to $1.6 billion per year.

That by itself makes the sugar program richly deserving of termination. However, additional details in the GAO report reveal a scale of dysfunction that should place the program firmly in Congress’s crosshairs. The following are six key points raised in the report:

The program significantly raises the cost of sugar in the United States.

As can be seen from the GAO chart, the price of raw sugar in the United States is significantly higher than abroad. In recent years it has been twice as expensive as raw sugar in other countries.

US sugar producers receive substantially more government support than their foreign counterparts.

Although the US sugar industry alleges “massive subsidization” by leading sugar‐​producing countries such as Brazil and India and claims to be merely trying to “survive in a world of heavily subsidized sugar,” the GAO report makes clear that the United States is among the most egregious practitioners of such subsidies. Indeed, the report shows that the United States is surpassed only by China in its enthusiasm for sugar subsidies as measured by the industry’s percent of revenue from government support.

The GAO report also points out that, beyond government measures explicitly devoted to supporting the US sugar industry, the business further shields itself from foreign competition via the use of antidumping and countervailing duty laws. As the report states:

Both the U.S. sugar program and the U.S. antidumping and countervailing duty suspension agreements protect the domestic sugar industry. This makes sugar an outlier in terms of trade protection; even before accounting for the suspension agreements with Mexico, sugar had by far the highest trade protection of any U.S. good, agricultural or nonagricultural, according to a 2017 USITC study.

The sugar program boosts the profits of sugar farms. Using both a previous GAO study and other studies, as well as data from the US Department of Agriculture (USDA), the report calculates that the US sugar program benefits each sugar farm by approximately $100,000 to $190,000 per year (although it cautions that the amount will “vary substantially” by farm size). These sugar farms, the report adds, are “substantially more profitable than non‐​sugar farms” to the tune of 24.3 percent higher net income per acre for farms growing sugarcane (compared to the average non‐​sugar farm) and 54.2 percent higher net income per acre for the average farm growing sugar beets.

Given the higher acreage of sugar farms compared to non‐​sugar farms, this means the overall net income of farms growing sugar beets was over seven times higher than non‐​sugar farms while farms growing sugarcane saw profits over eight times higher.

Furthermore, this is based on data from 2017. But from 2017 to 2022, the report adds that sugarcane prices rose by 44 percent and sugar beet prices rose by 48 percent while total US farm production expenses rose by 26 percent. In other words, the profitability of sugar farms may have further increased.

The poor disproportionately bear the burden of paying for the sugar program. The GAO report points out that the country’s lowest‐​income households spent an average of 30.6 percent of their income on food in 2021 compared to 7.6 percent for the highest‐​income households. Therefore, policies such as the sugar program that raise the cost of food—and there are plenty of foods that use sugar as an ingredient—are disproportionately felt by the poorest Americans.

The sugar program undermines US food manufacturers. Unsurprisingly, the GAO found that costly sugar is harmful to food manufacturers who rely on sugar as an input for the products they make. While sugar is a small cost for many food producers, it averages as much as 9 to 10 percent of material costs for the makers of candy and chocolate and, as the GAO notes, likely also harms other industries where sugar is an important ingredient, such as cookies and breakfast cereals (among others).

But the harm goes beyond just higher costs. Due to US restrictions on sugar supply, simply getting access to adequate amounts of sugar can be another issue for food manufacturers. That’s no small thing. According to representatives of sugar‐​using industries contacted by the GAO, sugar supply issues have, at times, “forced some confectioners to temporarily shut down production, cancel orders, or pay double the typical price of sugar.”

This helps explain why some food manufacturers have shifted production to other countries where they can be assured a ready supply of sugar at competitive prices. That means fewer jobs in the United States. A 2015 study highlighted by the GAO found that the US sugar program depressed employment in food manufacturing by 18,500 workers.

The sugar program’s tariff rate quote is inefficient. To help keep supplies low and prices high, the United States employs a tariff‐​rate quota that allows limited amounts of raw and refined sugar to enter the country each year under a relatively low tariff and forces all other imports to pay a tariff that is significantly higher. While the tariff‐​rate quota for refined sugar sets aside certain amounts for imports from Canada and Mexico and administers the remainder on a first‐​come, first‐​served approach to imports from all other countries, raw sugar imports use a more complex system that divides the tariff rate quota among 40 countries.

One big problem identified by the GAO report is that many of these quota assignments go unused. Of the forty countries receiving tariff‐​rate quotas, two have never used their quota, seven have not done so in the last fifteen years, and twenty have filled less than 75 percent of their allocations in total between 2006 and 2022.

While the USDA has the power—which it often exercises—to both reallocate and increase the tariff‐​rate quota during the year, significant problems still present themselves. As the GAO report states:

However, in 5 of the 11 years for which reallocations occurred between 2010 and 2022, the reallocations were announced in June or later. According to some tariff‐​rate quota‐​holding countries, this timing—three-quarters of the way through the quota year—occurs after they have already shipped their base allocation. Furthermore, these countries stated the timing makes it difficult to impossible to arrange the additional shipment on short notice.

Additionally, some sugar users we spoke to told us the timing of the reallocation and increase process has caused supply chain issues and has prevented sugar users from obtaining sugar needed for production. Specifically, they mentioned that when raw sugar is imported late in the fiscal year, sugar refineries are typically operating at or close to full capacity. As a result, the reallocated and increased raw sugar can take months to be refined for production, limiting the amount of sugar available, according to sugar users.

Every year dating back to 1996, raw sugar imports have been less than the in‐​quota quantities established by USDA, leading to persistent shortfalls.

From 2006 through 2022, the report points out that the United States imported an average of 13 percent less raw sugar than the in‐​quota quantities that had been set during those years. That percentage, the GAO adds, works out to about 2.8 million metric tons over that period with an estimated value of $1.67 billion. That’s money that was effectively left on the table due to the tariff‐​rate quota’s inefficient nature.

Summing up

Cato scholars have been pointing out the costly and nonsensical nature of the US sugar program for years. This latest GAO report only further underscores what has long been known. With the 2018 farm bill set to expire this year, the time is ripe for Congress to take a fresh look at US agricultural policy. Repealing this sugar industry giveaway should be high on any list of potential reforms.

Jeffrey A. Singer

Yesterday, Ohio voters approved Issue 2, which legalizes recreational marijuana, by a decisive 57 to 43 percent vote. Issue 2 allows adults aged 21 and older to possess up to 2.5 ounces of marijuana publicly and to grow up to six plants at home. Ohio is set to become the 24th state to legalize recreational marijuana. Thirty‐​eight states (including Ohio) and the District of Columbia have already legalized medicinal marijuana.

One potential fly in the ointment is that Issue 2 was a voter‐​initiated state statute, not a constitutional amendment. Therefore, under Ohio law, state legislators may rewrite or override much of the new statute.

Another potential problem is that the new statute imposes a 10 percent state excise tax on marijuana sales in addition to the state and local sales taxes, which will still apply. Ohio state and local sales taxes average around 7.24 percent. This means that people shopping at legal marijuana retailers may pay more than 17 percent tax on purchases.

Experience in California and other states that have legalized possessing and using marijuana should teach policymakers that overtaxing retail marijuana only serves to preserve the pre‐​legalization black market, as marijuana consumers who no longer fear arrest and punishment will pay less if they buy the weed from underground dealers.

Uruguay was the first country in the world to legalize marijuana production, sale, and consumption. Several European countries have decriminalized marijuana, and a number of European countries are contemplating legalizing it.

Now that the majority of the US population lives in jurisdictions that have legalized recreational marijuana use, it is long past due for Congress to act. Congress should pass federal legislation that makes growing, processing, transporting, selling, and using marijuana legal and leaves it up to states to decide on marijuana’s legal status for medicinal or recreational use within their borders.

But federally legalizing marijuana while it remains on the Drug Enforcement Administration’s schedule of controlled substances doesn’t go far enough. The DEA currently categorizes marijuana as a Schedule 1 drug: “no currently accepted medical use and high potential for abuse.” If the DEA re‐​scheduled marijuana to one of the remaining schedules (2 through 5), the plant would remain a controlled substance, requiring a prescription from DEA‐​licensed health care practitioners, whose prescribing patterns would be closely monitored and regulated by law enforcement.

Congress should require the DEA to de‐​schedule marijuana, i.e., remove it from the schedule of controlled substances. Alcohol, a much more potent and potentially harmful drug, is not on the DEA schedule of controlled substances. Neither should marijuana be. De‐​scheduling marijuana will also make it easier to do clinical research on the plant’s medicinal uses.

Surveys show bipartisan support for legalizing marijuana. A recent Gallup poll found overall support at 70 percent (87 percent of Democrats, 69 percent of Independents, and 55 percent of Republicans). As is often the case, the people are way ahead of the politicians. It’s time for leaders in Washington to jump out to the front of the parade line and wave the banner for marijuana legalization.

Marc Joffe

The Census Bureau recently reported that the US has 90,837 local governments. Many of these entities are superfluous, insolvent, or poorly managed. In several states, an oversight body is responsible for identifying, righting, and even dissolving local governments that are failing. For example, the North Carolina Local Government Commission dissolved the town government of East Laurinburg due to a chronic inability to manage its finances. The town is now an unincorporated community with government services provided at the County level.

Unlike North Carolina, California does not empower an executive agency to dissolve local governments, but it does give them oversight responsibilities. Both the State Controller and State Auditor have the power to conduct local government audits.

But with thousands of local governments across the Golden State, determining which ones to investigate is challenging. The decline of local newspapers means that fewer reporters are looking for cases of waste, fraud, and abuse, so reviewing local government financial statistics may be the best option.

In 2019, the California State Auditor implemented a new platform to identify underperforming local governments. The office collected their audited financial statements, extracted key statistics, and calculated ten financial metrics for each entity. Finally, the auditor calculated a composite financial health score for each government, choosing from among the lowest scoring for full audits.

The auditor published its scores on a public website called the Local Government High‐​Risk Dashboard. This site provided remaining local journalists and concerned citizens a way to quickly see how their city compared to others across the state. While the site would have ideally included all local governments, it was confined to cities and towns.

One city that came to the auditor’s attention through this analysis was El Cerrito, in the San Francisco Bay Area. In a March 2021 report, the auditor concluded, “The city’s poor budget development practices have led to substantial increases in budgeted expenditures that it cannot afford.” Since then, the city has implemented a corrective action plan under the auditor’s supervision. In a March 2023 review, the auditor found that “El Cerrito has improved its budgeting processes to provide meaningful information for making fiscally sound decisions.”

While the auditor operated its dashboard, local government fiscal health improved largely because of higher real estate values (which, in turn, drive up property tax collections) and federal support provided under the American Rescue Plan Act. As a result, fewer cities are exhibiting severe financial distress and thereby becoming audit candidates. These tailwinds in local government finance are tapering off in 2023 as entities spend down their federal aid and property values stabilize (or, in the case of offices, decline). The result may be a new crop of audit candidates.

Unfortunately, the State Auditor abandoned its dashboard in October, citing a need to divert staff to overseeing California’s state financial audit, which has been late for five consecutive years. In announcing the discontinuation of its local government high‐​risk dashboard, the office did not mention any plan to eventually reinstate the project, so it may be gone forever.

And if government finances are deteriorating, the California state government will lack a useful tool to monitor trends. This could leave policymakers open to surprises like the one that happened in 2012, when three California cities filed for bankruptcy in relatively quick succession.

Operating the dashboard required a substantial resource commitment at the State Auditor’s office. Because there is no single public repository for local government audited financial statements, employees have to spend considerable time locating the audits. And, because the audits are published in PDF format, finding and entering key financial statement metrics is unnecessarily time‐​consuming. Federal legislation intended to make state and local government financial statements “machine readable” should ameliorate this problem if these laws are properly implemented.

Hopefully, the California State Auditor will reconsider its decision to suspend local government financial monitoring efforts. North Carolina, with its comprehensive monitoring and dissolution powers, offers a template for what is possible.