I recently heard a legislator discuss potential budget savings that Missouri could see with cuts to its corrections system. To put the idea of these cuts into perspective, I looked for spending data and found this interview with Joseph Eddy, budget director for the state corrections department. An edifying excerpt:

The department’s Joseph Eddy says it costs $44.68 a day per inmate. He says $12.14 a day is for medical and mental health services. Another $2.54 pays for three meals a day.

Prisoners also are paid for their work—7-dollars-50 cents a month. If they have their GED, they can earn an extra dollar. That’s about 35 cents a day.

Eddy says the direct costs of each inmate every day is $16.39 a day. The other $28 go for the administrative and prison personnel, utilities, and other costs that go with running a prison.

Per my calculations, that amounts to $16,308.20 per year, which is 26 percent smaller than the federal poverty line of $22,050 — and less than I would have expected. A quick search of the per-inmate expenditures in peer states suggests that Missouri ranks toward the bottom in its per-inmate expenditures.

It is interesting that the direct costs of each inmate amount to only 37 percent of the total cost per each inmate. Even more interesting is that medical care represents a full 74 percent of the direct costs of each inmate. To me, all of this suggests that there is actually not a lot of fat that can or will be trimmed, and that those looking to prioritize significant savings would do well to look elsewhere. That being said, there may still be ways for the state corrections department to operate more efficiently. In a subsequent post, I’ll discuss how Missouri can create markets and design incentives in ways that can save it some of the money it currently spends on corrections.

I am writing this post from India, where I have been vacationing the past few weeks. One of the most striking features of the social organization here is the prevalence of private schooling. Selection bias aside, I have yet to meet a student who attends a public school. It is not uncommon for impoverished children and children of slum residents to spurn free public education in favor of private education. Recognizing this, I have been wondering how the increased share of private school students here affects levels of educational innovation, test scores, and other educational outcomes. Fortunately, I’ve just found an interesting new study that evaluates this effect. The authors, Martin West and Ludger Woessmann, purport to measure the treatment effect of the “share of schools that are privately operated” on PISA (Programme for International Student Assessment) scores. Of course, there is a problem here with determining causality. As they write:

Countries where more people choose to invest in private schools may have other attributes, such as higher income levels or a greater commitment to education, that lead to better achievement. If this is the case, any positive correlation between private schooling and student achievement could reflect a country’s income or educational commitment rather than any beneficial effects of competition. Or it may be the case that low-quality public schools increase the demand for private schooling. If so, then it could seem that competition lowered public school quality when in fact the causal connection could be in the opposite direction.

To solve for this, West and Woessmann exploit a technique that economists call an instrumental variables approach. Essentially, they found a new variable — the size of a country’s Catholic population in 1900 — which is a useful predictor for the current share of schools that are private. Because the new variable is a good predictor of the variable they are really interested in (share of schools that are private), but is presumably not influenced by higher income levels or greater commitment to education, using this new variable is more useful for determining causality.

Their results (emphasis added):

Our results indicate that the share of schools that are privately operated has an economically and statistically significant positive effect on student achievement in mathematics, science, and reading, even after controlling for the current levels of Catholics and for the share of funding that privately operated schools receive from the government. Larger historical Catholic shares that translate into a ten percentage point larger private school sector today increase average student achievement on the math test by 9% of an international standard deviation. Science and reading achievement increase by roughly 5% of a standard deviation. These patterns are evident despite the fact that the contemporary share of Catholics in each country is negatively related to student achievement, suggesting that distinctive cultural features of traditionally Catholic countries are unlikely to be driving these results. Importantly, much of the positive effect of private school shares accrues to students in public schools, suggesting that the overall effect is not simply due to privately operated schools being more effective, but rather it reflects benefits of competition.

It’s not clear how applicable these results are to private school competition within only the United States or within individual states such as Missouri. However, if the same pattern holds for educational institutions in Missouri, then policies that encourage private school competition, like vouchers or tuition tax credits, will have spillover benefits for public school students as well — not just the recipients of educational aid.

By the way, those readers who are now beginning to suspect that I spend most of my free time perusing the economics of education literature would be correct in that suspicion.

A newly released report evaluating the impact of Milwaukee’s voucher program (Milwaukee Parental Choice Program) provides evidence that the program has improved student outcomes. The study notes that prior evaluations have demonstrated positive effects of the voucher program on test scores:

Do students benefit by using the Milwaukee Parental Choice Program (MPCP) to attend a private school instead of a Milwaukee Public Schools (MPS) school? In addressing that question, most prior evaluations focus on whether students in the MPCP score better on tests of academic achievement than students in MPS schools. As reviewed in our 2007 report, two studies based on randomized trials each demonstrated significantly higher mathematics test scores for MPCP students as compared to MPS students four years after enrolling in the program; one study also showed significantly higher reading test scores.

In contrast, the focus of the present study was to investigate the effect of the voucher program on graduation rates:

Overall, had MPS graduation rates equaled those for MPCP students in the classes of 2003 through 2009, the number of MPS graduates would have been about 18 percent higher. That higher rate would have resulted in 3,939 more MPS graduates during the 2003-2009 years. A recent analysis of the economic impact of high school dropouts suggests that the annual impact from an additional 3,939 MPS graduates would include an additional $24.9 million in personal income and about $4.2 million in extra tax revenue.

An important consideration to bear in mind when reviewing education research is that educational goals are varied, and a wide variety of educational outcomes can be implicated by policy changes. As such, even if a policy change is demonstrated to have a negligent impact on an important student outcome such as test scores, that policy may still positively impact other desirable student outcomes. In the case of the voucher program in Milwaukee, it appears that several student outcomes are positively implicated by school choice. In other regions of the country, school choice programs have had a much less significant impact on test scores. It is important that researchers and the general public then probe the effects of such programs on other desirable outcomes, as well, before a summary judgment of the policy is made.

A recent working paper from the National Bureau for Economic Research adds to the evidence that “performance pay” for students produces gains. From the abstract:

Policymakers and academics are increasingly interested in applying financial incentives to individuals in education. This paper presents evidence from a pay for performance program taking place in Coshocton, Ohio. Since 2004, Coshocton has provided cash payments to students in grades three through six for successful completion of their standardized testing. Coshocton determined eligibility for the program using randomization, and using this randomization, this paper identifies the effects of the program on students’ academic behavior. We find that math scores improved about 0.15 standard deviations but that reading, social science, and science test scores did not improve.

The Coshocton program is funded with money contributed by a local businessman.

Although 0.15 standard deviations may seem small, it’s not bad compared to other educational interventions. In terms of improvement per dollar spent, this program seems more efficient than other more popular interventions, like lowering class sizes.

A few weeks ago, I testified at the Missouri Tax Credit Review Commission’s meeting in Columbia. I’d like to highlight one specific point from this speech.

One particular member of the commission (I do not remember which one) attacked a previous speaker who had recommended that all tax credits be abolished. The commission member suggested that since every other state uses tax credits, Missouri’s exit from the business of providing tax credits would put us at such a disadvantage that could result in Missouri no longer producing anything. There are many possible responses to this complaint that tax credit opponents can employ; for brevity, I will note just one.

Comparative Advantage:

Consider two states: Missouri and California. Suppose that these two states have firms that can produce two goods: wine and computers. Now, suppose that firms in Missouri can produce six bottles of wine or three computers per hour, whereas firms in California can produce 12 bottles of wine and four computers per hour. In this case, California firms have a higher productivity and we would say that California has an absolute advantage in the production of both wine and computers. This does not, however, imply that California will, or should, produce both goods.

One of the key insights from introductory economics courses is that comparative advantage matters. Instead of evaluating productivity in terms of outputs, we can evaluate productivity in terms of opportunity cost. Note that, in this example, when a Missouri firm produces one bottle of wine, it misses an opportunity to produce half of a computer. Similarly, when a Missouri firm produces one computer, it misses an opportunity to produce two bottles of wine. We can think of these missed opportunities as costs. For California firms, the cost of producing one bottle of wine is a third of a computer, and the cost of producing one computer is three bottles of wine. So, our example shows that even when California has an absolute advantage in the production of both goods, Missouri still retains a comparative advantage in the production of computers because its opportunity cost (two bottles of wine) is lower than the opportunity cost for California firms (three bottles of wine). Thus, in this limited illustration, it would be more efficient for Missouri to produce computers and trade with California for wine.

We can apply this insight to tax credits. Suppose that California aggressively courts winemakers and computer manufacturers with tax incentives and Missouri does not. One way to think about these incentives is that they work to lower the marginal costs that firms face, which allows a firm to produce more. This makes it appear as though California firms are more productive in translating inputs (in dollars) into outputs (in product volume). As our example illustrates, even if these apparent increases in productivity give Californian firms an absolute advantage in the production of certain goods, it is likely that Missouri will still retain comparative advantage and will continue to produce many of the goods that California chooses to subsidize.

If you’re like me, you too have spent the past few months eagerly awaiting the release of a certain academic article. Well, I’m pleased to report that the wait is over. Here it is: “How Does Your Kindergarten Classroom Affect Your Earnings? Evidence From Project STAR.” The paper is written by several “superstar” economists whose research supports the conclusion that investments in improving the quality of early education can provide lasting benefits. The paper is also noteworthy for tracking the effects of early educational interventions onto market outcomes, and not just subsequent educational outcomes.

Here’s the abstract:

In Project STAR, 11,571 students in Tennessee and their teachers were randomly assigned to different classrooms within their schools from kindergarten to third grade. This paper evaluates the long-term impacts of STAR using administrative records. We obtain five results. First, kindergarten test scores are highly correlated with outcomes such as earnings at age 27, college attendance, home ownership, and retirement savings. Second, students in small classes are significantly more likely to attend college, attend a higher-ranked college, and perform better on a variety of other outcomes. Class size does not have a significant effect on earnings at age 27, but this effect is imprecisely estimated. Third, students who had a more experienced teacher in kindergarten have higher earnings. Fourth, an analysis of variance reveals significant kindergarten class effects on earnings. Higher kindergarten class quality – as measured by classmates’ end-of-class test scores – increases earnings, college attendance rates, and other outcomes. Finally, the effects of kindergarten class quality fade out on test scores in later grades but gains in non-cognitive measures persist. We conclude that early childhood education has substantial long-term impacts, potentially through non-cognitive channels. Our analysis suggests that improving the quality of schools in disadvantaged areas may reduce poverty and raise earnings and tax revenue in the long run.

For those who don’t want to read the whole paper, the research is also available as a PDF of Power Point slides.

This old article from the Sacramento News & Review contains some interesting sentences about sub-prime credit:

While the Chicago Outfit may have been a bit heavy-handed in its debt-collection practices, the interest rate the crew charged for a loan was a bargain. A bargain, that is, compared to the fees charged by the numerous payday loan outfits in Sacramento and throughout the state.

Carlisi and company extended short-term credit, or “juice loans,” for fees that pencil out to an annual interest rate of 260 percent. The Outfit may be disappointed to learn that they were working for chump change. Had they waited a few years, and then come out West, they could have become payday lenders and made some real money.

Although the gratification of physically collecting a loan isn’t allowed, in California it’s perfectly legal for a state licensed payday lender to charge up to 5,474 percent annual interest in this rapidly expanding niche lending business.

I’ve been meaning to comment on this for a while, because this is really fascinating data. Readers who peruse the article from which this excerpt is lifted will note that the author uses this statistic to argue that payday rates are excessive and exploitative. Well, perhaps, but this data doesn’t render that claim obvious. The fact that payday loan rates are higher than loan shark rates could simply suggest either that payday lenders face higher costs of enforcement, higher default rates, higher transaction costs, lower-quality information, or some combination of these factors.

It’s easy to see how a legitimate, white-market business would have higher overhead costs than a black market loan scheme, if for no other reason than that a white-market business must handle contractual disputes with tools furnished by the legal environment. No such encumbrances burden black market creditors. As former Show-Me Institute Policy Analyst Justin Hauke put it in an op-ed: “At least with a payday lender, default is settled in court. In the black market, it usually involves a crowbar.” In this sense, the higher prices of payday loans likely reflect the premium that consumers are willing to pay for safety.

From USA Today, “Merit pay study: Teacher bonuses don’t raise student test scores”:

Offering middle-school math teachers bonuses up to $15,000 did not produce gains in student test scores, Vanderbilt University researchers reported Tuesday in what they said was the first scientifically rigorous test of merit pay.

Some 296 middle-school math teachers — two-thirds of the district’s middle-school math teachers — volunteered to participate in the experiment. Half were placed randomly in a control group, while the rest were eligible for bonuses of $5,000, $10,000 or $15,000 if their pupils scored significantly higher than expected on the statewide exam known as the Tennessee Comprehensive Assessment Program.

Except for some temporary gains for fifth-graders, though, their students progressed no faster than those in classes taught by the 146 other teachers.

The PDF for the study is available online.

As a merit pay advocate, I’d love to disparage these results as the product of some unsound methodology, but I can’t, in good faith, do that. This seems like a relatively clean experiment. Yet merit pay supporters need not abandon their cause. The study provides good answers, but the questions may be too narrow to be fully relevant. For example, in evaluating the responsiveness of teachers to potential performance bonuses, the study approximates what a labor economist would call elasticity of effort but not elasticity of labor supply. Put differently, the study suggests that the performance of existing teachers may not change in the presence of performance incentives but the study does not consider the dynamic changes in the overall teaching pool that may result from implementation of merit pay programs. More research must be conducted to evaluate whether merit pay attracts a better pool of educators who, in turn, have positive impacts on student performance.

Saint Louis’s Paideia Academy, a charter school, is set to close its doors following a recent defeat in a court battle with Missouri’s Board of Education, which rejected the school’s charter application earlier this year. The Post-Dispatch reports that the Board of Education, in rejecting the application, and the Cole County Circuit Judge, in upholding its decision, cited poor management, the lack of a sponsor, and low test scores as reasons to revoke the charter. Although I am not in a position to speak about the quality of management, or about the lack of a sponsor (which certainly seems like a valid reason to revoke a charter), I do, however, object to the “low test score” argument on two grounds.

First, although it is true that Paideia’s test scores rank among the lowest in the state, absolute measures of test scores are not a very meaningful measure of school quality. The production of education is similar to the production of anything else in the economy: Poorer quality inputs, in the form of poorer students from historically disadvantaged ethnic backgrounds, translate to poorer quality outputs, in the form of test scores. It’s not only a mistake, then, to compare Paideia’s students to those of high-performing districts, but also to an arbitrary benchmark determined by the state. Taking a snapshot of test scores is not enough, because a reliance on mere glimpses into time discourages an understanding of the underlying trends at work. The more important measure is the longitudinal one: Are Paideia’s students learning more now than they were before the school existed? Perhaps the answer is no, but it doesn’t look like this question was considered by either the Board of Education or the Cole County Circuit Judge.

Second, I am willing to believe that we may overvalue test score measures of all kinds. One-size-fits-all models don’t work in schools, where abilities and interests vary greatly between student populations. Schools that produce less significant test score gains but more significant “creativity” gains may still be cultivating meaningful human capital.

Consider: There are more payday loan storefronts in the United States than there are McDonald’s and Starbucks outlets combined. Also consider, these payday loan storefronts are much more geographically concentrated than other types of outlets. Whereas Starbucks and McDonald’s sprawl across disparate locations with very unique compositions and characteristics of residents, payday storefronts tend to cluster densely in regions where demand for payday loans is likely to be high. What do these conditions imply about the characteristics of the payday loan market?

For starters, basic economic intuition would suggest that the payday lenders operate in a competitive marketplace. Fairly low barriers to entry (both legal and financial) into the market and the vast number of storefronts implies that individual stores face strong incentives to underprice their competitors. The result, barring collusion or market distortion, would be that prices are efficient, and not exorbitant.

The empirical evidence bears out this claim. A paper released by the FDIC Center for Financial Research used panel data from a large vendor to demonstrate that, despite the high interest rates on payday loans, the profitability of payday lenders does not statistically differ from the profitably of other financial intermediaries, like “reputable” banks. This should appeal to intuition: Payday lenders cater to risky populations that are vulnerable to financial stressors and prone to defaults. Risky customers warrant high rates to compensate for high default rates. This understanding regarding the level of market competitiveness and the condition of interest rate efficiency is crucial to understanding the policy effects of regulation in the payday loan market.

Last week, in a conversation with state Sen. Mary Still — one of Missouri’s most vocal critics of the payday lending industry and author of regulatory legislation in the General Assembly — I hoped to identify her latitude of acceptance for various payday lending policies (including deregulating the market further). I discovered that the two policy tools that are most likely to hear debate in the General Assembly are interest rate caps and providing incentives for banks to become “legitimate” vendors of payday loans. In some important ways, these approaches are troubling. If the market is already competitive and interest rates are efficient, an interest rate cap will choke the market and force lenders out — and banks shouldn’t have the ability to offer significantly cheaper rates on similar products. At any rate, revealed preferences would suggest that there is a reason banks aren’t willing to offer payday loans without incentives.

As I’ve discussed earlier, payday loans have the potential to be both helpful and harmful. Imposing interest rate caps on the market will stifle the ability of payday loans to help consumers, and incentivizing banks to offer such loans will do little to shield consumers from harm.

The Springfield News-Leader today features a good op-ed about current plans for regulating Missouri’s payday loan industry.

Good bit:

The FDIC found that payday loan fees were justified by the costs and risks associated with offering such loans. The FDIC also found competitive products like bounced checks carrying APRs of up to 3,500 percent.That APR calculation – designed to compare competing, long-term forms of credit – is why a 36 percent APR cap, as proposed in current Missouri legislation, would ban short- term loans in the state.

If imposed, a 36 percent rate cap would mean lenders could only charge about $1.38 per $100 borrowed. At such a low rate, lenders simply can’t cover their costs – such as rent, employee salaries and benefits.

As I’ve written before, I’m opposed to payday loan regulation because:

1. I view payday loan transactions as legitimate, consensual business interactions between relatively rational actors.
2. The empirical evidence suggests that payday loans constitute a useful service. I look, for example, to Donald Morgan and Michael Strain, who show that increased access to payday loans reduces the volume of bounced checks. I also look to Edward Lawrence and Gregory Elliehausen, who find that payday loans “satisfy a real financial need within a certain segment of the population.” As I cite these authors, I’m fully willing to concede that there is literature out there that disagrees with their claims. The reading list I composed earlier lists some of those papers. In a future blog post, I will attempt a more detailed comparison of the methodologies employed in the different studies.
3. If payday loans are useful, then limiting or eliminating the payday loan market will drive consumers to underground or black markets. This is not favorable, for reasons that should be self-evident.
4. I think the most legitimate critique of payday loans is that it disadvantages the politically weak who have, for example, little access to legal recourse. If that’s the case, the better solution would be to reform the political/legal apparatus, rather than the payday loan market. Opponents can argue that this is less feasible, and they would be correct, but if the market is driven underground, then these people would have no legal recourse anyway.

My main concern now is the third. Those who seek to regulate payday loans toe a narrow line between tempering the market and hobbling it. Unfortunately, it looks as though the proposed reforms are poised to do the latter.

One problem with the debate over payday loan regulation in Missouri and elsewhere is a lack of sustained focus on data. Regrettably, both opponents and proponents of regulatory legislation within the state seem to cling reflexively to familiar, abstract narratives and consequently fail to engage the public with meaningful evidence to support their assumptions. To alleviate this problem, I am compiling this list of literature — both sympathetic and unsympathetic to the payday loan industry — to enrich the public dialogue. If any of you know of more quality literature on the topic, please add to this post in the comments.

1. Payday Holiday: How Households Fare after Payday Credit Bans (ungated), Donald P. Morgan and Michael R. Strain.

   “Compared with households in states where payday lending is permitted, households in Georgia have bounced more checks, complained more to the Federal Trade Commission about lenders and debt collectors, and filed for Chapter 7 bankruptcy protection at a higher rate. North Carolina households have fared about the same. This negative correlation—reduced payday credit supply, increased credit problems—contradicts the debt trap critique of payday lending.”

2. The Economics of Payday Lending (ungated), John P. Caskey, Swarthmore College.

   General overview of payday lending industry and basic issues. Written for a lay audience.

3. Do Payday Loans Cause Bankruptcy? (ungated), Paige Marta Skiba and Jeremy Tobacman.

   “Though the size of the typical payday loan is only $300, we find that loan approval for first-time applicants increases the two-year Chapter 13 bankruptcy filing rate by 2.48 percentage points.”

4. Factors Affecting the Location of Payday Lending and Traditional Banking Services in North Carolina (ungated), Mark L. Burkey and Scott P. Simkins.

   Explores the geography of payday loan institutions. “A key finding is that after controlling for many covariates, race is still a powerful predictor of the locations of both banks and payday lenders.”

5. The Profitability of Payday Loans (ungated), Paige Marta Skiba and Jeremy Tobacman.

   “Despite charging effective annualized rates of many thousand percent, we find lenders’ firm-level returns differ little from typical financial returns. The data are consistent with an interpretation that payday lenders face high per-loan and per-store fixed costs in a competitive market.”

6. Quantifying the Economic Cost of Predatory Payday Lending (ungated), Keith Ernst, John Farris, Uriah King:

   “Our analysis of quantitative data reveals that payday lenders collect the vast majority of their fees from borrowers trapped in a cycle of repeated transactions, where borrowers are forced to pay high fees every two weeks just to keep an existing loan outstanding that they cannot afford to pay off.”

7. A Comparative Analysis of Payday Loan Customers (gated), Edward C. Lawrence and Gregory Elliehausen.

   “By analyzing the data collected in a national survey of payday customers, this research allows policymakers to better understand what type of consumer borrows from payday lenders, for what purpose, and what the true benefits and costs are. The results confirm a strong demand for payday loans that satisfy a real financial need within a certain segment of the population.”

8. Mayday Payday: Can Corporate Social Responsibility Save Payday Lenders (ungated), Carmen M. Butler and Niloufar A. Park.

   “In this article we ask what the best ways are to maximize the wealth of the payday lending industry while limiting the industry’s harmful impact on consumer communities? We assert that payday lenders will likely demonstrate greater corporate social responsibility only after there is a change in the laws that govern the industry coupled with industry-wide reform in corporate governance.”

9. Restricting consumer credit access: Household survey evidence on effects around the Oregon rate cap (ungated), Jon Zinman.

   “Borrowing fell in Oregon [after interest rate caps] relative to Washington, with former payday borrowers shifting partially into plausibly inferior substitutes: bank overdrafts and late bill payment. Additional evidence suggests that restricting access caused deterioration in the overall financial condition of Oregon households. Overall the results are consistent with restricted access harming, not helping, consumers on average.”

10. Consumers’ Use of High-Price Credit Products: Do They Know What They Are Doing? (gated), Gregory Elliehausen:

    This paper asserts that consumers of payday loans are sufficiently rational. A caveat, however, is that rationality is a just a process and does not imply that “good” decisions are made.

Some op-eds include:

• In Defense of Usury (gated), by Dean Karlan and Jonathan Zinman
• Congress Takes Aim at Payday Loans (ungated), by Robert DeYoung
• Payday Loan Reform Bad for Borrowers (ungated), by Justin Hauke

The last of those op-eds was written by a former employee of the Show-Me Institute. Perhaps unsurprisingly, my views on payday loans are fairly similar to his. Taking an economic view, I’m concerned that regulatory reform will be unable to limit payday loan harms effectively without driving the market underground. Taking a political view, I view payday loan consumers as sufficiently rational and believe that a government (at least in this arena) has more of an imperative to maintain free, private contracts than to protect the politically weak.