Medicaid Restrictions Could Impact Infant Mortality

If you’re involved in the public health world in Ohio, you’ve certainly heard about the problems the state has with infant mortality, with the state ranking in the top ten states for infant death rate according to the CDC. At the same time that Ohio struggles with infant mortality, federal regulators have made news by authorizing states to block grant a portion of Medicaid dollars, potentially reducing access to Medicaid coverage by low-income adults. State lawmakers have been skeptical of adopting such a plan at this point, though the legislature does have a track record of restricting access to the program in the past.

It should be noted that federal officials have stated pregnant women will not be impacted by the block-grant scheme. That being said, limiting access to Medicaid for non-pregnant women could make it more difficult for a pregnant woman to understand when she is eligible. Also, there is some evidence that states with Medicaid expansion have shown progress on infant mortality despite pregnant women not being a part of the expansion population, suggesting that either the policy is making it easier for pregnant women to care for their children or that some other factor is at work in expansion states that is leading to reductions in infant mortality rates.

This study is one of the latest in a long list of studies establishing a link between Medicaid coverage and infant mortality. While experimental evidence of Medicaid’s impact on physical health is famously lacking, quasiexperimental evidence of the nation’s largest safety net program’s impact on infant mortality has been mounting for the past twenty-five years. One study suggested higher Medicaid payments have significant impacts on infant mortality rates while another on Medicaid eligibility suggested just being eligible for Medicaid had an impact on child mortality. A later study found higher Medicaid payments could reduce incidence of low birthweight, a significant risk factor for infant mortality. Another study found that expansions of Medicaid access can lead to lower fetal death rates and a more recent study found access to care can reduce racial gaps in infant mortality.

The fact that most of these studies deal specifically with pregnant women as a population suggests that exempting pregnant women from policies that could restrict access could save lives for children down the road. But restrictions for other populations could have spillover effects as well as women find it harder to identify when they are eligible. Care should be taken to craft Medicaid policy that lines up with state public health goals because, as this body of evidence shows, access to care can impact the lives of infants.

Ohio Minimum Wage Hikes Would Likely Increase Wages, Decrease Employment and Poverty

The minimum wage conversation has come to Ohio. Earlier this week, Ohio Attorney General Dave Yost certified ballot summary language for a proposed Ohio constitutional amendment to raise the state’s minimum wage from the current $8.70 to $13 per hour. This proposal, backed by state service worker and teachers unions, comes at the same time that Democrats in the Ohio House are pushing a $15 minimum wage.

So what would happen if Ohio adopted a higher minimum wage? Would it lead to higher wages and reduction in poverty as advocates for the policy claim? Or would it lead to an increase in unemployment as detractors of minimum wage hikes argue?

One place we can look for guidance is an impressive study by the Congressional Budget Office (CBO) on proposals to increase the federal minimum wage. The CBO used Current Population Survey data and available empirical data on employment elasticities to estimate the wage, poverty, and employment impacts of minimum wage hikes.

A quick and dirty way of using this data is to scale the scope of this federal study down to the size of the state of Ohio. There are some reasons this approach is limited: the CBO study includes some states that have higher minimum wages and some that have lower minimum wages than Ohio’s, which falls in the middle of the distribution of states. Ohio is a tad older, a bit whiter, and notably poorer than the country as a whole. A full study replicating CBO’s methodology would be valuable. Unfortunately, I only have the time to write a blog post about this, so I will stick with the imperfect scaling model, reducing CBO’s numbers to 3.6% of projected people affected and assuming linear impacts to extrapolate the impacts of a $13 minimum wage until someone will (shameless plug!) pay me to make more precise estimates. While this approach is not perfect, the direction of the impact is likely to be the same and overall takeaways similar (undermining my plug) to a more thorough replication.

Below is a chart of the projected number of people who would be impacted by the two proposed minimum wage increases based solely on CBO projections scaled to the size of the state of Ohio followed by a discussion of what conclusions we can draw from them.

Data from the Congressional Budget Office and the US Census Bureau.

Data from the Congressional Budget Office and the US Census Bureau.

Raising the Minimum Wage Will Likely…Raise Wages

This may seem obvious, but a higher minimum wage means higher wages for Ohioans—potentially a lot of Ohioans. Using this methodology, we estimate that over 320,000 Ohioans could see their earnings increase in an average week under a $13 minimum wage and almost 610,000 Ohioans could see their earnings increase in an average week under a $15 minimum wage. This represents the number of Ohioans who are currently working under the proposed minimum wage thresholds.

Similarly, 280,000 Ohioans are estimated to be just above the $13 wage threshold and 670,000 are just above the $15 wage threshold, putting them in a range that could be impacted by employers raising wages to comply with the new wage levels. This means that a total of 600,000 to 1 million Ohioans could experience higher wages under a $13 or $15 minimum wage scenario. This total amounts to 11-17% of the total workforce of the state in December 2019.

Wage increases would likely be concentrated among the poor and near-poor as well. Assuming wage increases grow linearly from $12 to $15 and that increases in Ohio would look similar to those reported in the CBO report, a $13 minimum wage would increase an Ohio family in poverty’s wages by 2.7% and a $15 minimum wage would increase its wages by 5.3%, putting hundreds of dollars in the pocket of the average poor family in Ohio. Families in the 100-300% of poverty (making on average $29,000 to $56,000 a year) would also benefit from the increase, families in the 300-600% of poverty range (averaging $95,000 a year) would be marginally impacted, and families making over 600% of the federal poverty level (averaging $230,000 a year) would see wages decrease.

Data from the Congressional Budget Office.

Data from the Congressional Budget Office.

A Minimum Wage Hike Will Probably Reduce Employment

All increases of $12 an hour and higher estimated by the Congressional Budget Office led to a median estimate of a net decrease in employment. While an increase in the minimum wage could bring workers with higher reservation wages into the workforce and could increase employment by correcting for labor market power imbalances slanted towards employers, the CBO projects those impacts would not be as large as the classic supply and demand dynamics that lead to reduced employment in the face of higher wages.

It is worth noting that CBO’s projections all include negligible impacts to employment in its likely range of outcomes, encompassing two-thirds of potential outcomes. It is also worth noting that the high end of this range is quite substantial, with likely estimates skewed towards higher employment impacts. Thus, more extreme employment impacts exist on the high end of the estimate than the low end. Using the same methodology we used to estimate number of workers with wage impacts, we can estimate that a $13 minimum wage would lead to a reduction of 0-75,000 jobs in Ohio with a median estimate of 17,000 lost jobs and that a $15 minimum wage would lead to a reduction of 0-130,000 jobs with a median estimate of 29,000 lost jobs. For reference, Ohio added about 110,000 jobs in 2019, which means that a $13 minimum wage would cost 0-71% of 2019 job growth with a median estimate of 16% and a $15 minimum wage would cost 0-130% of 2019 job growth with a median estimate of 27%. The highest likely impact of a $12 minimum wage would be to lose 1.3% of the total December 2019 workforce size and the highest likely impact of a $15 minimum wage would be to lose 2.4% of the total December 2019 workforce size.

Data from the Congressional Budget Office and the US Census Bureau.

Data from the Congressional Budget Office and the US Census Bureau.

Minimum Wage Increases Will Likely Reduce Poverty

Lastly, the CBO estimates 400,000 Americans would be pulled out of poverty by a $12 minimum wage and 1.3 million Americans would be pulled out of poverty by a $15 minimum wage. Scaling this down to the size of the state of Ohio and assuming a linear trend in poverty reduction from $12 to $15, a $13 minimum wage in Ohio would pull 25,000 Ohioans out of poverty and a $15 minimum wage would pull 46,000 Ohioans out of poverty. In 2018, 1.6 million Ohioans were in poverty, meaning a $13 minimum wage would lead to a 0.2 percentage point decrease in the poverty rate and a $15 minimum wage would lead to a half percentage point decrease in the poverty rate.

It was stressed above and it is worth stressing again: the precision of these estimates could be honed by a more thorough investigation of state-level data. That being said, the takeaways that a minimum wage would likely raise wages for a substantial portion of the low-income workforce, somewhat reduce employment, and have a modest impact on poverty would probably hold even with a more sophisticated analysis. Now it is up to policymakers and the general public to evaluate those tradeoffs and determine their taste for the levels of projected increased wages and decreased poverty in exchange for projected decreased employment.

Scioto Analysis Partnering with UC Berkeley on Genuine Progress Indicator Study

This Spring, Scioto Analysis will be partnering with UC Berkeley’s Goldman School of Public Policy on an update to its November 2018 study of the Ohio state economy. The research team will use the Genuine Progress Indicator framework, a “GDP+” measure that combines traditional economic indicators with environmental and social indicators, as a baseline for economic growth then will explore policy options for policymakers interested in growing the state economy effectively, efficiently, and equitably.

“This partnership is exciting because it will extend the findings of our 2018 study past descriptive statistics and into the territory of policy analysis,” said Scioto Analysis Principal Rob Moore. “The point of this measure is not only to inform policymakers of the state of the economy, but also to provide guidance to policymakers who want to improve it.”

The research team consists of the following researchers, all graduate students in the Master of Public Policy Program at UC Berkeley’s Goldman School of Public Policy.

  • Cruz Eduardo Flores Vera, Economic Analyst, who previously worked for the Central Bank of Mexico doing analysis of energy markets.

  • Masashi Hamano, Environmental Analyst, who previously worked for the Ministry of Finance in Japan as a budget analyst.

  • Isabel Clayter, Social Analyst, who previously worked in financial technology and consumer lending in San Francisco, focusing on consumer finance.

  • Ashwin MB, Environmental/Social Analyst, who most recently worked for the Abdul Latif Jameel Poverty Action Lab in New Dehli, India studying governance and transportation interventions.

The research team plans to focus on GPI measurement over the next month then turn to policy analysis in March and April.

Housework and Parenting in Ohio is a $92 Billion Industry

Scioto Analysis estimates that the value of housework and parenting in the state of Ohio in 2016 came out to about $92 billion. For reference, this total amounted to 29% of the state Genuine Progress Indicator (GPI), a “GDP+” measure that includes environmental damage and social indicators such as the value of higher education and the cost of lost leisure time next to traditional economic indicators.

If we compare this estimate to data on other industries in the state of Ohio that are measured to calculate GDP, we can see that housework and parenting would be a larger industry than any other in Ohio besides manufacturing if it was measured as such. The value of housework and parenting is 28% larger than the state professional and business services industry, 33% larger than both the state real estate/rental/leasing industry and the government industry, and is two-thirds larger than both the finance/insurance and health care industries.

Data from the Bureau of Labor Statistics, American Time Use Survey, and Bureau of Economic Analysis

Data from the Bureau of Labor Statistics, American Time Use Survey, and Bureau of Economic Analysis

Scioto Analysis calculates the value of housework and parenting using standard methodology for calculation of state GPI. First, we use American Time Use Survey data to estimate how many hours a day the average adult spends on household activities and caring for and helping household members. We then multiply these numbers by average housekeeping and child care wages respectively to determine what the daily value of these activities would be in the form of market wages. These are then multiplied by the number of people in Ohio age 15 and up to reflect the population engaging in these activities and then multiplied by 365 to convert from individual daily value to statewide annual value.

From an efficiency standpoint, this does not tell us a lot. What it says is that $92 billion of activity is generated by workers who are paid nothing for what they do. This sort of analysis does not tell us whether these activities have social benefits (though at least for childrearing it would be surprising to find that that they don’t), just that people are doing a lot of unpaid work.

From an equity standpoint, standard economic theory would suggest that low-income people will be more likely to engage in more hours of housekeeping and caring for household members since they have less opportunity to make high wages in the market. Thus, providing supports for people engaging in these activities may be a way to deal with inequities in human capital. Andrew Yang has made such arguments when championing UBI.

A big reason this number matters, though, is that raising children is central to what many people consider “the good life.” Surveys suggest that child-related activities score higher than all leisure activities besides sex on self-assessed enjoyment among women and men. While economic development strategies that increase employment may lead to higher economic output as measured by GDP, they may be neutral or even harmful to well-being when factoring in the second-largest industry: housework and parenting.

Does Fade-Out Fade Out?

Early childhood education has a problem in the research community, and it’s called “fade-out.” If you work in early childhood research, you’ve heard this term before. If you haven’t, the general problem is that some research suggests that gains from early childhood education programs realized by children as they enter kindergarten “fade away” in elementary years to the point where they are nothing by third grade. If this is true, then early childhood education programs could be expensive programs that ultimately yield little results for children or families.

The savvy policymaker, though, will ask the following question: why should I care about third-grade test scores? Well, third-grade test scores are nice because they allow us to evaluate a program after a couple of years rather than decades. The problem with third grade test scores is that they might be the least predictive results to extrapolate to life outcomes.

Timothy Bartik, a leading economist of job creation, has studied the impact of early childhood education on local wages. In his book From Preschool to Prosperity: The Economic Payoff to Early Childhood Education, Bartik tackles the question of fade-out. He looks at the four most high-profile studies in early childhood education: experimental studies the Abecedarian Project and the Perry Preschool Project, and quasiexperimental evaluations of Head Start and the Chicago Child-Parent Center Program.

As can be seen above, estimates of adult earnings based on third grade test scores are below the estimates of adult earning effects at end of the early childhood program. Thus, Bartik does find evidence for fade-out. However, Bartik also finds evidence of a strong bounce-back from fadeout, with actual adult earnings higher not only than third grade test scores would predict, but also higher than the original end-of-program test scores would have suggested.

These results are also reflected in cost-benefit results reported in a high-profile literature review by the Rand Corporation. The review reported the results of 12 evaluations and meta-analyses of early childhood education programs, some broken out into treatment group subcategories. While the evaluations that only included results from elementary school tended to show negative results, evaluations that followed up with participants in secondary school, early adulthood, and middle adulthood showed increasing net benefits.

Data from Karoly et al., Early Childhood Interventions: Proven Results, Future Promise, pp.xxvi-xxvii

According to Rand,

The largest benefit-cost ratios were associated with programs with longer-term follow-up because they allowed measurement at older ages of outcomes such as educational attainment, delinquency and crime, earnings, and other outcomes that most readily translate into dollar benefits (p. xxv).

This suggests that third grade scores could be missing latent human capital that was built by early childhood scores such as socialization, emotional intelligence, or verbal communication skills that then end up leading to future education, crime reduction, and labor market outcomes.

Ultimately, third grade test scores have little social relevance on their own: they are only a proxy for future outcomes with immense social importance: educational attainment, crime victimization, and labor market earnings. While third grade test scores let us more quickly assess the impact of policies, if these scores are not correlated with socially-relevant outcomes, they are not of much use to policymaking. Maybe the problem isn’t the policies: it’s the limitations of how we test their impacts.

Three Education Programs that Work

The state of the American education system can be pretty discouraging. Despite the challenges levied inequities in achievement, uneven funding, and disparities in graduation rates, though, many policymakers are still striving to increase the number of students in poverty experiencing academic success. But where should they start?

One place to look is the Washington State Institute for Public Policy’s Benefit-Cost Database. We’ve used WSIPP to look at workforce development strategies before, but the state cost-benefit analysis database also has information on interventions that range from criminal justice to health care to higher education. Here, we look at pre-K through 12 interventions.

These results need to be taken with a grain of salt. While the methods are rigorous, they are prepared for policymakers in Washington State, so the policy may operate differently in a place like Ohio. That being said, they still give us some guidance for policies that be used to improve academic outcomes for students in poverty. Below are the three programs that have the highest per-participant net benefits in WSIPP’s education policy database.

Becoming a Man (BAM) with High-Dosage Tutoring

Becoming a Man” is a cognitive behavioral therapy-based behavioral program focused on skill training and exposure to prosocial adults for disadvantaged high school-aged young men. This program is conducted in a one-hour weekly group sessions where participants learn character and social-emotional skills such as considering others’ perspectives, evaluating consequences ahead of time, and reducing automatic decision making. WSIPP’s analysis evaluated the impacts of Becoming a Man combined with daily, two-to-one ratio hourly math tutoring sessions.

While the program costs $4,700 per participant, it yields over $40,000 in per-participant benefits due to improved test scores and increased labor market earnings associated with them. This means that not only does the program yield net social benefits, but the increased wages garnered from the program will lead to increased tax revenue that will pay for the cost of the program in the long run. This also means that a federal (or even a state) government would be incentivized to fund such a program for budgetary reasons alone.

The drawbacks of Becoming a Man are typical. Like any educational intervention, the payoff takes time. WSIPP estimates that about 12 years will pass before the benefits of this program exceed the costs. This is faster than some other educational programs since its participants are high-school age, but still some time to wait. Additionally, the program has some uncertainty to it. Since this program has only been evaluated in limited circumstances and benefits hinge on the program leading to test scores which lead to labor market earnings, there is some uncertainty about whether they will pan out. In the limited evaluation of the program that has been carried out, though, the results were so positive that WSIPP is optimistic about future applications of this model.

Consultant Teachers: Literacy Collaborative

While Becoming a Man focuses directly on students, the Literacy Collaborate model indirectly impacts students by coaching teachers. Literacy coaches are trained for up to 35 days at local universities then provided with ongoing training support after placement in schools. Coaches work one-on-one with teachers to improve instructional practices. The evaluation WSIPP used focused in particular on the impact of the model on classrooms with children in early primary years, grades K-2.

Literacy Collaborative costs are very low, coming out to only $780 per student according to an Ohio State University study of the program in Columbus, Ohio cited by WSIPP. This means the program costs one-sixth the cost to operate as Becoming a Man. This is likely because of the “teach the teacher” model, where less resources can be used to impact more students. Benefits are also large, at about $28,000 in new future labor market earnings per child impacted, less than Becoming a Man but still substantial. This means it is also a program that “pays for itself” down the road.

This program has one similarity to Becoming a Man and one glaring difference. The similarity is that the program takes a long time to pay off, though even longer than Becoming a Man at an estimated 16 years before the program yields net benefits. The difference is that the WSIPP Monte Carlo simulations used to predict the probability the program has net benefits showed net benefits in 100% of simulations. This means that the program is a slam-dunk to provide benefits if carried out correctly in the right environment. This is likely because of the program’s low cost: as long as it can cover the $780 per student, it will yield net benefits.

Double-Dose Classes

The final of these three programs is deceptively simple: double-dose classes. If a student is struggling in math or reading, that student will be enrolled in two math or reading classes instead of one, thus doubling their instructional time.

It is surprising how effective this simple intervention is. At the low cost of only about $500 per participant (marginal costs of another student in another class are fairly low), about $18,000 in new future labor market earning benefits are accrued. In addition, there is a small impact on crime from the program as it has been found to reduce crime rates as well. Because of its low cost and the future labor market benefits, the program also pays for itself in the long run.

Double-dose classes have certainty of yielding net benefits of near the same level as Literacy Collaborate, with 98% of simulated scenarios resulting in a net-benefit program. Additionally, double-dose classes yield benefits in about ten years, quicker than Becoming a Man or Literacy Collaborative.

These are only three of the 42 educational programs WSIPP has determined yield net benefits, so it really only scratches the surface of tools policymakers have to improve academic outcomes for students in poverty. What I take away from this, though, is that there is no reason to get cynical about K-12 education in the United States. We have tools that work: we just need to use them.

9 Studies that Rocked the Policy World in 2019

As 2019 comes to a close, it’s a good time to reflect on the state of policy research in the past year. 2019 was a big year for policy research, with impactful studies coming out of diverse institutes such as NBER, the CBO, and the National Academies of Sciences, Engineering, and Medicine. Below are nine studies that particularly stood out in over the past year in informing policy debates.

National Academies: A Roadmap to Reducing Child Poverty

In December 2015, Congress passed an appropriation measure that included a provision for the National Academies of Sciences, Engineering, and Medicine to research and present policy options for cutting the American child poverty rate in half in the next ten years. Three years later, the panel of 15 leading poverty researchers released a detailed, 600-page study of child poverty in the United States, along with microsimulated policy packages designed to reduce child poverty in the United States.

A conservative, $9 billion plan was proposed by the study committee. The package of tax credit, minimum wage, and job training policies would reduce child poverty about one fifth, making it the most cost-effective package the panel came up with though it still fell short of the 50% child poverty reduction goal. To hit that goal, the committee concluded the federal government would have to spend $90-110 billion, utilizing either a mix of tax credits and housing and food assistance expansions or a package of tax credits, minimum wage increases, and a combination of child allowance, child support assurance, and immigration liberalization measures.

CBO: Key Design Components and Considerations for Establishing a Single-Payer Health Care System

In May of 2019, the Congressional Budget Office released its long-awaited report on single-payer health care. In characteristic CBO fashion, the study rankled partisans on both sides, revealing the complexities of redesigning the U.S. health care system from scratch while declining to project costs of any particular proposal.

This study was particularly important because of the shortage of dispassionate, evenhanded analysis of single-payer proposals in the United States. There is plenty of reading to do on single-payer in the U.S., but opinions are usually marred by rosy predictions of extreme reductions in spending or ideological opposition to state participation in the health care markets. This CBO report lays out how single-payer health care operates in the six countries that currently have a single-payer systems and walks policymakers through twelve questions they will have to answer if interested in designing a single-payer system. This report should be required reading for any advocate or detractor of single-payer health care as well as in undergraduate or graduate level courses on the U.S. health care system.

Moody’s: Stress Testing States 2019

This study is unique to this list in a couple of ways. First, it is the only study on this list carried out by a firm owned by a publicly-traded corporation. Most corporations like this are less focused on creating non-revenue generating policy relevant research than centers like CBO and NBER. Second, this is the only entry on this list that is a regular study put out on an annual or semiannual basis. Though the methodology of this study is not new, it is rigorous and is particularly relevant as economists debate how close the next recession is and state policymakers debate how big their nest egg should be when this recession hits.

In this October 2019 study, Moody’s Analytics does the hard work that state budget offices usually decline to do, estimating what potential shortfalls in state revenue and shocks to budget needs will look like under different recession scenarios. While this report was covered rather rosily, with many outlets reporting that a majority of states are prepared for a moderate recession, Moody’s still found a number of states that would require belt tightening, especially under conditions of a severe recession. Both fiscal conservatives and advocates for the poor should read this study if they want to understand the tradeoffs between fiscal sustainability and poverty alleviation we make in current budgets and what decisions are kicked down the road to future state budgets.

NBER: Shrinking the Tax Gap: Approaches and Revenue Potential

Studies out of the National Bureau of Economic Research often speak less to headline-grabbing policy problems than those in the CBO or the National Academies. That being said, the beauty of work published in NBER is that it unearths key policy problems that impact lots of people underneath the surface of current policy debates and authors are often encouraged to put forth bold policy proposals to solve these problems.

This November 2019 study is penned by an unlikely pairing of young Penn Law Professor Natasha Sarin and Larry Summers, as big a star as you get in the world of economics. The study addresses a decidedly unsexy topic of uncollected taxes and proposes an equally unsexy policy package of audits, increased reporting requirements, and IT modernization to increase revenue collection by the IRS. What’s sexy are the numbers: Sarin and Summers estimate that these low-cost administrative changes could raise $1 trillion in extra revenue over the next ten years, enough to finance a myriad of new programs or tax rate reduction packages, including being able to finance on its own the most expensive policy packages for reducing child poverty in that National Academies report listed above.

CBO: The Effects of Tariffs and Trade Barriers in CBO’s Projections

An addendum to economic projections might not seem like a anything worth writing home about, but CBO Economist Daniel Fried’s August 2019 explanation of the impact of the trade war on the U.S. economy was a classic example of the Congressional Budget Office playing the part of the “skunk at the company picnic.” Fried reported that tariffs imposed by the Trump Administration in 2018 reduce the size of the U.S. economy by 0.3%, costing the average U.S. household about $580. Also characteristic to CBO, the explanation includes a couple of paragraphs about the uncertainty of these estimates. Overall, the 90-page economic outlook update mentions the word “tariffs” 124 times, interpreting trade policy as a significant factor in U.S. economic growth in the upcoming years.

NBER: A Market for Work Permits

This month, leading international poverty economist Martin Ravallion teamed up with former World Bank colleague Michael Lokshin to put forth an ambitious proposal for breaking the stalemate around high-skill work permits in the United States. Lokshin and Ravallion acknowledge that citizenship and residency barriers create a de facto “entitlement” to work in a certain country based on accident of birth. Provocatively, the two economists ask what would happen if citizens interested in pursuing nonmarket pursuits such as caring for children could sell their right to work in the country to immigrants itching for opportunity to use their skills in the United States. Loshkin and Ravallion estimate that a seller of a permit could make $16,000 in the first year of the program and that poverty would be reduced by a third by the infusion of new resources from selling of work permits and tax revenue from new migrants’ earnings. The impacts are projected to fade as immigration markets stabilize over time, but Loshkin and Ravallion still project impacts to be substantial ten years down the road, providing thousands of dollars to sellers of work permits and continuing to substantially reduce poverty.

Upjohn: Making Sense of Incentives: Taming Business Incentives to Promote Prosperity

Timothy Bartik is one of the most prolific and policy-relevant economists in the field of business incentives and in the policy world in general. His work developing cost-benefit models particular to the goal of raising local wages has illuminated the world of business incentives and has been instrumental in understanding the potential of well-designed early childhood programs in promoting local economic development. In his October 2019 book, Bartik pulls together the literature on business incentives for the age of Amazon HQ2.

Like any good policy researcher, Bartik puts his money where is mouth is, laying out a framework for an “ideal” state business incentive program focused on getting the largest local wage impact bang for incentive buck invested. Bartik suggests targeting economically stressed counties with slack employment capacity and high-tech counties likely to generate larger local economic development benefits. He suggests creating a basis of infrastructure and workforce development services, then following up with customized business services, and finally creating limited, state-financed, up-front tax incentives open to all firms in targeted areas.

CBO: The Effects on Employment and Family Income of Increasing the Federal Minimum Wage

Once again, the CBO is not afraid to wade into choppy policy waters, this time estimating the employment and income impacts of proposals to raise the federal minimum wage to $10, $12, and $15. Again, the CBO rankles those on the left and right by projecting minimum wage increases would have substantial impacts on employment while also reducing income inequality and boosting average wages for families in and near poverty. Notably, the July 2019 report finds that a $10 minimum wage would have negligible effects on both employment and incomes while a $15 minimum wage would boost average family wages for those in poverty by 5% at the likely cost of over 1 million jobs nationwide. Also notable are the range of employment impacts: CBO projects as little as no employment impact and as much as 4 million jobs lost under a $15 minimum wage scenario.

NBER: How Research Affects Policy: Experimental Evidence from 2,150 Brazilian Municipalities

In this study, which Scioto Analysis has written about before, U.S. economists team up with Innovations for Poverty Action to evaluate how policymakers respond to policy information. In two studies of municipal policymakers in Brazil, researchers find that not only do policymakers apply policy research when they are exposed to it, but they also have a willingness to pay for this research. While Scioto Analysis’s existence is a testament to this fact, it is exciting to see the lesson learned in a large-scale experimental setting. With better evidence comes better policy, and only through partnerships between researchers and policymakers will policy be able to ultimately be improved.

Here’s to a great 2019 in policy research and hope for an even better 2020!

Workforce Development Strategies Can Alleviate Poverty If Done Right

In Harry Holzer’s book Where Are All the Good Jobs Going?: What National and Local Job Quality and Dynamics Mean for U.S. Workers, the Georgetown economist argues that “good” jobs aren’t going away—they’re just changing in character. In particular, good-paying jobs now require more analytical communication skills than in the past and thus are being filled by more highly skilled workers.

Policymakers interested in reducing poverty have thus been interested in finding ways to improve the skills of low-income residents of their community. Unfortunately, many job training programs have not been found to be effective as reported in a Council of Economic Advisers study earlier this year. So what strategies are available for local policymakers interested in effectively increasing access to training for residents experiencing poverty?

One tool we have to answer this question is the Washington State Institute for Public Policy (WSIPP), the most sophisticated cost-benefit analysis institute in the country. WSIPP has been commissioned by the Washington State legislature to study over three hundred different public policy interventions and to estimate the monetized impact of each of these interventions. While WSIPP’s estimates are prepared for policymakers in the state of Washington and thus have limitations for extrapolation to other contexts, they still can be informative for policymakers in other states considering workforce strategy.

Below is an overview of the per-participant net benefits of the ten workforce development programs WSIPP has analyzed. The programs WSIPP has studied fall into four major categories: career academies, job search and placement, case management, and training/work experience. As can be seen below, WSIPP has found significant differences both between and within these categories of interventions. Let’s take a look at each of these categories one by one.

Career Academies

Career and technical education academies have been described as “school within a school” high school programs that foster connections between schools and local employers. They are designed to build both vocational and academic skill sets for students to apply in workforce and postsecondary settings. WSIPP analyzed the results of studies of California’s Career Academies and Linked Learning program for its cost/benefit estimates.

These estimates suggest that career and technical education academies yield the highest net social benefits of all the workforce interventions studied, amounting to nearly $9,600 in net benefits per participant. Benefits mainly accrue from improved labor market earnings associated with employment, about two-thirds of which accrue to participants and one-third accruing to taxpayers. Career and technical education academies also yield $2.70 per $1 invested in the program, so provide an attractive tool for policymakers working under a firm budget constraint.

The drawback of these programs is that benefits accrue over a long time horizon, estimated to take almost a decade for the program to break even, compared to instant payoff for job search and placement and case management programs and 3-5 year payoffs for training/work experience programs. Career and technical education academies are also fairly expensive, costing an estimated $5,700 per participant, ten times the cost of job search and placement programs and thirty times the cost of low-cost case management programs. This means that career and technical education programs would cost more to assist the same number of people, though the return on that cost would be larger.

Job Search and Placement

Programs that focus on job search and placement are brief interventions lasting from a few hours to two months comprised of supervised job search, job search workshops, or job clubs for unemployed individuals. These programs are usually targeted towards unemployment insurance claimants or TANF (welfare system) participants.

Job search and placement programs yield $2,000 in benefits per participant and $4.71 per dollar invested in the program, making them moderately cost-effective from a net benefits perspective and very cost-effective from a cost/benefit ratio perspective. Benefits also accrue immediately compared to the 8-year lag for career and technical academies and are cheap at about $500 per participant.

The drawback of these programs is that benefits mainly accrue to taxpayers, with taxpayers enjoying nearly three times the benefits of program participants between increased tax revenue and decreased public assistance spending. While the average participant in a job search and placement program walks away with $1,100 in increased labor market earnings, she also loses $400 in public assistance, decreasing the effectiveness of the program as an antipoverty tool.

Overall, job search and placement programs are quicker but less effective tools than career and technical academies. That being said, they represent a better tool than a status quo of inaction since they do yield benefits for both participants and taxpayers.

Case Management

Case management services usually comprise counseling and job search and retention services and referrals to work supports such as child care subsidies, transportation, education, and training. Case management can be conducted via orientations, assessments, interviews, or telephone calls in individual or group sessions.

Interestingly, WSIPP’s estimates for the cost effectiveness of case management hinge strongly on the characteristics of the participants. Unemployed individuals benefit greatly from case management, accruing $2,700 in labor market benefits at the cost of only $200 per participant. Current or former welfare recipients, in contrast, only only accrue an average of $200 in labor market earning benefits at a much higher cost of about $3,000 per participant.

The takeaway from these findings is that case management programs can be quite effective, but only if targeted towards unemployed individuals and not current or former welfare recipients.

Training/Work Experience

Training and work experience are what we usually think of when we talk about workforce development. These services may comprise job search and placement assistance, adult education, English as a second language, GED, vocation training, child care, transportation, and paid and unpaid jobs. Programs can start with training and then move to work experience, consist of an individualized employment plan from an employer, or just consist of job training or work experience.

Work experience on its own is a moderately effective intervention, yielding a net of $1,600 in labor market earnings per participant after subtracting benefit reductions. Adding training to a work experience program makes the benefits per participant higher (a net of $5,100 after subtracting benefit reductions), but the higher costs associated with training lead to lower net benefits than work experience on its own. Targeting work experience/training interventions towards welfare recipients can lead to higher social yields, though, due to decreased spending on public assistance. This makes this sort of targeting attractive to state and federal policymakers, but potentially less attractive to local policymakers who have less to gain from reductions in public assistance spending.

Targeting work experience and training interventions towards youth yield substantial net costs to the tune of over $10,000 per participant. This is because these interventions are especially costly when targeted towards youth and have very low yields to the tune of only about $150 per participant. Training without work experience has large labor market benefits per participant of $6,000 per participant, but are incredibly costly, making the approach much less attractive than work experience-focused interventions.

Takeaways

Overall, WSIPP’s analysis leaves us with some good guidance for policymakers looking to reduce poverty through workforce interventions. First, career and technical academies are incredibly effective but long-term investments in workforce. Policymakers looking for cheaper, quicker interventions can invest in case management specifically targeted towards unemployed individuals and job search and placement programs. As far as training/work experience programs go, work experience is a key ingredient and should generally be supported, except when targeted towards young people, who do not tend to benefit from such programs. A mixture of short- and long-term interventions could be a good formula for improving skills to reduce poverty.

How Local Governments Can Create High-Paying Jobs for Residents

How do we create living wage and high-paying jobs for residents of neighborhoods of concentrated poverty? One tool we have are business incentives, or, as leading national business incentive economist Timothy Bartik describes them, “tax breaks, cash grants/loans, or services that are (1) targeted at an individual firm, or some industry or group of firms, and (2) intended to promote job growth in a state, or in a local geographic area that is big enough to be a local labor market.”

Business incentives are a (some may say “the”) key tool for creating good jobs at the local level, but economists estimate that anywhere from 75-98% of business incentives have no impact on the decisions of firms to relocate, expand, or retain workers.

In light of this discouraging evidence, local policymakers are right to ask how they can use business incentives more effectively. One of the things that makes Bartik a leader not just in business incentives but in policy research in general is his willingness to put his money where is mouth is and lay out proposals for better policy on the topic he studies. In his recent book Making Sense of Incentives: Taming Business Incentives to Promote Prosperity, Bartik lays out his proposal for an “ideal state incentive program.” While this is tailored to state governments, lessons can be gleaned for local governments as well.

Target distressed areas with high unemployment

Jobs in distressed areas are more likely to go to local residents. Bartik suggests targeting counties, but at the local level, counties and municipalities may want to target certain zip codes or census tracts based on unemployment, poverty levels, or other indicators of economic distress. In particular, looking at objective measures that areas lack adequate jobs can be good guidance for local governments.

Start with basic services supporting economic development

Bartik suggests prioritizing general economic development services over funding services for specific businesses, specifically mentioning infrastructure and high-quality programs for skills development. This is a suggestion that can be directly adopted by local government. Neighborhoods of concentrated poverty often have poor infrastructure, making them unattractive for business. Investment in road and utility infrastructure can improve the ability of these neighborhoods to attract business. These neighborhoods also tend to have lower education levels, so skill training services provided or subsidized by city or county agencies could help improve workforce prospects in neighborhoods and provide partners for businesses interested in locating or expanding in these neighborhoods.

Next, prioritize funding for customized business services

Bartik suggests block granting state funds to counties, but at the local level, these funds could probably be managed by a development agency. He suggests funding be spent on services for “tradable industries”. This means that industries should compete in state or national markets: funds spent on firms that compete locally just shift jobs around the local market rather than bringing new business to the region.

Bartik mentions the following examples of customized business services: manufacturing extension services, small business development centers, business incubators, customized job training, and discretionary hiring subsidies for firms that hire into newly created jobs local nonemployed residents referred and placed via local workforce agencies. All these are strategies that can be used by local governments, especially small business services, customized job training, and hiring subsidies.

Lastly, Bartik suggests a level of funding for these customized services that would provide quality services to all targeted firms. His suggestion is $10 billion nationally. Scaled down on a per-person basis to an example jurisdiction like Franklin County, that would come out to $40 million spent on customized business services, which amounts to about 9% of the county budget—a significant number. For reference, raising this amount of new funds from county sales taxes, the broadest tax the county raises, would require an increase in the county sales tax rate from 1.25% to about 1.4%.

Make tax incentives limited in costs, up front, and open to tradable firms of all sizes

It should go without saying that tax incentives should not exceed the amount owed to local government, but having an explicit policy can keep costs from spiraling out of control. Bartik also suggests making payments up front and using clawbacks to enforce performance measures. This is because businesses value present dollars greater than future dollars, so incentives are more effective if they are front-loaded rather than accruing years down the road. Bartik also recommends making incentives a legal entitlement or having specific provisions that level the playing field between smaller and larger firms.

These are just four suggestions, but the takeaway is this: business incentives can be done better with smarter targeting, more attention to infrastructure and customized services, and prudent policy around tax incentives. Strategies such as these can pay off for residents of neighborhoods of concentrated policy.

Mapping Neighborhoods of Concentrated Poverty in Franklin County

Mapping technology has contributed a lot to understanding the impact of geography on public well being. Luckily, you don’t have to be a GIS whiz to work with geographic data. The US Census Bureau provides a pretty solid tool for analyzing geographic disparities.

For instance, let’s look at the way poverty is distributed throughout Franklin County, Ohio, using only Census Bureau tools. Below you can see a map of poverty prevalence in Franklin County generated using the Census Bureau’s tools. To generate this map, I simply used the American Fact Finder’s search for census tract five-year average poverty rates from 2017 then chose an arbitrary point - 25% poverty, and split the census tracts accordingly. I played with a few different poverty thresholds and tried three classes and four classes, but I think this two-category split best shows the concentration of poverty within the county.

Figure 1. Concentrated Neighborhoods of Poverty in Franklin County

Figure 1. Concentrated Neighborhoods of Poverty in Franklin County

Zooming in, we can see which neighborhoods have the most concentrated poverty. Overall, this visualization suggests six neighborhoods of concentrated poverty: Downtown/Near East Side, the East Side, the South Side, the West Side, Linden, and Ohio State Campus.

Downtown/Near East Side

Figure 2. Concentrated Poverty in Downtown/East Side

Figure 2. Concentrated Poverty in Downtown/East Side

Besides a little pocket in the King-Lincoln district, every census tract from the Scioto River to Alum Creek and between I-70 and I-670 is over 25% poverty. This may seem surprising to people talking about the “gentrification” of what is coming to be called “Old Towne East” along with the rapid building of downtown condos, but poverty is still quite persistent on the Near East Side and downtown. These numbers may change, though, over the coming years. Since this data lags back to 2012 in order to increase sample size, there is a possibility we will see this area start to turn more green in the coming decade.

East Side

Figure 3. Concentrated Poverty on the East Side

Figure 3. Concentrated Poverty on the East Side

Poverty on the East Side stretches from the Fifth Avenue neighborhoods north of Bexley to the neighborhoods east of James Road and west of Yearling Road in the Eastmoor/Whitehall area all the way down to the old Eastland Mall area. While the west side of Whitehall has tracts of concentrated poverty, the east side of Whitehall is notably better off.

South Side

Figure 4. Concentrated Poverty on the South Side

Figure 4. Concentrated Poverty on the South Side

South side neighborhood poverty is an interesting story. You can see the Brewery District, German Village, Schumacher Place, and Merion Village neighborhoods around Schiller Park are all under 25% poverty, but to the east in Southern Orchards and to the south in Hungarian Village, we still see concentrated poverty. We also see concentrated poverty in the Driving Park and Alum Creek Road areas to the southeast and even poverty stretching all the way down the Lockbourne Road area before it gets to Obetz and South High Street all the way down to the 270 loop. With development around Parsons Road, some are wondering if the Southern Orchards area will see declines in poverty rates, but poverty still seems to be persistent in the area for the time being.

West Side

Figure 5. Concentrated Poverty on the West Side

Figure 5. Concentrated Poverty on the West Side

Concentrated poverty on the West Side of Columbus is kind of a tale as old as time. Tracts of concentrated poverty predominate Franklinton and much of the greater Hilltop area and even the tract hugging the west side of the Scioto River opposite Marble Cliff. Notable are two tracts of concentrated poverty outside the 270 loop, signaling some suburban concentrated poverty on the West Side. A bright spot is the Westgate area, which is experiencing lower poverty rates than Greater Hilltop around it. Eyes will be on Franklinton as large development projects may impact local poverty rates in East Franklinton over the next few years.

Linden

Figure 6. Concentrated Poverty in Linden

Figure 6. Concentrated Poverty in Linden

Linden is another neighborhood that has garnered a lot of attention for its poverty. The area of concentrated poverty centered in Linden also spills over into the Northland, North East, and North Central areas. Many City of Columbus antipoverty projects have been centered in the Linden area, likely because of its geographic size and the extent of concentrated poverty in the area.

Ohio State

Figure 7. Concentrated Poverty in the Ohio State Campus Area

Figure 7. Concentrated Poverty in the Ohio State Campus Area

Probably the strangest neighborhood of concentrated poverty in Franklin County is that around the Ohio State University campus. As we’ve written before, college student concentration can impact poverty rates quite substantially, so much so that the Census Bureau has written briefs on how college students impact poverty rates. Dealing with college poverty can be deceiving because it is often (but not always!) temporary, volitional, and maybe even incorrect. Place-based poverty interventions in this neighborhood should likely take different forms than those in other neighborhoods due to the unique circumstances of poverty in a college area.

As you can see, some simple mapping can tell you some very interesting stories about poverty in a place like Franklin County. Using maps like this can help guide policymakers towards place-based policies to reconcile these geographic disparities. But place-based approaches need to be tailored towards specific neighborhoods. For instance, a keen eye may have noticed a dot of red on the southern border of Franklin County that didn’t get an analysis here. Well that tract covers the Rickenbacker National Guard Base, which means the “poverty” problem here is quite different from that in Linden or on the West Side. This type of mapping only gets you so far on its own: you need to marry this geographic data with an understanding of the local conditions each neighborhood is facing in order to craft smart place-based antipoverty policy.