A year or so ago, I met my friend’s mother for the first time at a wedding. She told me that she was Mississippi born and raised, but that after her kids were born she and her husband decided to move to North Carolina. Turns out the whole extended family was from Mississippi, still lives there, still loves it there.

“Why did you leave?” I asked.

“Because we had little kids, and the schools were terrible.”

Her answer didn’t surprise me – I’d heard about Mississippi’s bad schools before. But while its schools were terrible enough to induce a cross-country move when her kids (now in their mid-twenties) were young, that’s no longer the case. 

Mississippi has become an educational role model, a shining example of what’s possible inside public schools. It’s a turnaround story no one expected.

Mississippi is, on average, a state that people leave. It has the fourth-lowest in-migration rate in the country (only Louisiana, Michigan, and Ohio have fewer transplants from other states), while 36 percent of its young people move out-of-state. On net, its population is shrinking. Between 2020 and 2024, 16,000 more Mississippi residents died than were born.

Mississippi is a state known for its poverty, its unreliable infrastructure, and its substandard health care system – as well as its poor overall public health. It leads the nation in pregnancy-related deaths and high infant mortality rates. Its capital city, Jackson, has contamination issues with its water supply (with an annual average of 55 breaks per 100 miles of water line, nearly four times the national safety limit of 15). Mississippi consistently comes in as the poorest state in the country, with one in four Mississippi children living below the poverty line.

It’s not a state most Americans look to as a role model.

But over the past fifteen years, this unassuming Deep South state has been quietly pulling off one of the most impressive feats in American public education: while literacy rates around the nation have been falling, Mississippi’s have been steadily rising.

Historically, Mississippi’s school system performed about as well as its health care system and its economy: that is, near the bottom in the national rankings. For years, Mississippi ranked 50 out of 50 in the country for K-12 education. But all that changed in 2013, when Mississippi implemented the Literacy-Based Promotion and embraced the science of reading, overhauling its K-3 literacy curriculum and its teacher training.

Since 2013, Mississippi’s overall K-12 achievement scores have improved significantly. In 2013, Mississippi came in 49 out of 50 states on the NAEP (Nation’s Report Card) for fourth grade reading. In 2021, that number jumped to 21 – and in 2024, it rose all the way to ninth in the nation.

All of this was achieved while Mississippi faced a slew of challenges: teacher shortages, low teacher pay, and under-resourced special education programs, to name a few – the things critics so often point to as the culprits for poor educational outcomes. And all of this was achieved too in a state where 26-28 percent of its students are living below the poverty line – the children who are historically the most underserved (and therefore the lowest performing) students in the country.

All these challenges make Mississippi’s achievements more impressive, and the conclusion more irrefutable: reading science works. A measured, methodical, science-driven approach to teaching literacy results in – you guessed it – unprecedented levels of literacy.

That should not be a headline. And yet it is, printed and reprinted all over the country, colloquially referred to as “the Mississippi Miracle” – because the comeback story is so impressive, so unprecedented, so unexpected.

And yet, the strange thing isn’t that one of the poorest and most under-resourced states in the country implemented this – the strange thing is that it’s so rare as to be noteworthy.

Mississippi’s turnaround story is, as most things in life, a story of cause and effect – and in this case, the causes are quite few: a scientific approach to reading, a teacher education program consistent with that scientific approach, early identification and intensive intervention for students who are struggling, and a commitment to honoring the integrity of grade level standards (if a child isn’t reading at a third grade level, they don’t get advanced to third grade).

The “scientific approach to reading” in question is – no surprise – teaching via phonics, the time-tested approach to literacy that has worked for centuries, but which modern public schools seem strangely allergic to.

The simplest headline summary of the Mississippi Miracle is that Mississippi started teaching its kids to read using phonics – and stopped advancing kids who hadn’t learned the material. Their literacy scores turned around seemingly overnight. But of course, the story is more complicated than that.

Mississippi’s comeback started all the way back in 2000, in the private sector, when corporate executive and philanthropist Jim Barksdale donated $100m to launch the Barksdale Reading Institute, a nonprofit intended to turn around Mississippi’s poor literacy rates. Barksdale, whose résumé included serving as the COO of FedEx, the CEO of AT&T, and the CEO of Netscape, was deeply committed to his home state of Mississippi and deeply concerned about the literacy rates in its schools.

He saw the literacy crisis for what it is: the deficit of a fundamental life skill, with lasting implications for the entire life trajectory of children robbed of the chance to learn to read.

As sociology professor Beth Hess wrote to The New York Times after Barksdale’s donation was announced (after praising Barksdale himself): “It is disturbing that the state of Mississippi will be rewarded for its continuing failure to tax its citizens fairly and to allocate enough money to educate students, especially in predominantly black districts. This should have been a public rather than private responsibility.”

Yet as is so often the case, it was private sector efforts that led to change, unfettered by bureaucracy and untethered from the slow-moving weight of the public sector machine.

The Barksdale Reading Institute tackled the reading crisis at every level: teaching reading instruction inside Mississippi’s teachers’ colleges, engaging with parents and early childhood programs (like Head Start), and educating teachers on teaching phonics.

In 2013, Mississippi’s public sector followed suit, implementing two critical steps: passing a law that required all third graders to pass a “reading gate” assessment to advance to fourth grade, and appointing Casey Wright as Mississippi’s superintendent of education, who in the words of journalist Holly Korbey, “reorganized the entire education department to focus on literacy and more rigorous standards.”

Under the stewardship of Wright, Mississippi trained over 19,000 of its teachers in teaching phonics using the science-backed instructional program LETRS. In the early days of the literacy push, the state focused more on teacher training than on curriculum, but in 2016 it expanded its efforts to promote the use of curricula it felt best supported literacy training.

Compared with a full curriculum overhaul, the third-grade reading gate might sound like a small change, but it’s a critically important piece of the puzzle. Across the country, grade advancement is largely treated as a product of age, not of academic ability. Students with “failing grades” can be held back (and often are), but a passing grade is a low bar: a “D,” often considered a passing grade, usually means proficiency of 60 percent, meaning a child can miss 40 percent of the third-grade material and still advance to fourth grade.

The third-grade reading assessment ensures that children aren’t advancing to harder material with large gaps in their knowledge, that they’re set up with the skills they need to succeed, rather than being thrown in the deep end to fail. It’s also an important milestone: third-grade reading proficiency is a leading indicator of long-term academic success, with poor third-grade readers far more likely to drop out of high school. And as evidenced by Mississippi’s rising math scores (even though most of its energy is being directed toward literacy), the ability to read correlates with better performance across all subjects.

All this effort, unsurprisingly, led to swift and measurable results. Not only did Mississippi come in ninth in the nation in fourth-grade reading in 2025, but it scores even higher when weighted for demographic factors like poverty.

None of this should be scientifically surprising (because obviously teaching kids to read using the scientifically backed approach was going to work). But it’s politically shocking because, despite ample research, schools across the country resist teaching students to read using phonics, and their literacy rates flounder as a result.

Other states across the South (coined by Karen Vaites as the “Southern Surge states”) have followed Mississippi’s lead. Louisiana implemented a similar reading program in tandem with Mississippi, beginning in 2012 and seeing similar results. Tennessee implemented approaches borrowing from Mississippi and Louisiana in the school year of 2018-19, and Alabama followed suit in the 2019 legislative session. Each state is seeing success in its amended reading education approach.

None of these states have ample funding; each is in the bottom half nationally for per-pupil spending. All of these states have large numbers of students below the poverty line. Some have teacher and resource shortages. And yet, by implementing a pure phonics approach to reading instruction, they’re blowing past states that have more funding and more resources but are using a less rigorous approach.

In the words of writer Kelsey Piper, “illiteracy is a policy choice.” We know that teaching reading via phonics works. We know how to do it. And, thanks to Mississippi, we know it can be effective even with a limited budget and limited staff. 

Thanks to Jim Barksdale, we know that private sector pushes toward better policy can be effective. And thanks to states using non-phonics literacy approaches (and whose test scores are falling while Mississippi’s are rising), we know what not to do, too. 

The challenge now is to stop doing what doesn’t work, and start moving toward what does – not just in Mississippi and the Southern Surge states, but all across the country.

For most of us, especially those of us who think about it a lot, the Roman Empire conjures up famous names of such men as Caesar, Augustus, Nero, Marcus Aurelius, and a few others of the imperial elite. We might also think of grand structures like the Colosseum, the Appian Way, and the Pantheon, or massive spectacles from gladiator duels to races at the Circus Maximus. Dozens of books explore the Empire’s wars against Dacia in southeastern Europe, the Iceni in Britannia, Germania in northern Europe, and the Jews in Palestine.  

The point is, we tend to think of the extraordinary, not the ordinary or, to put it another way, the macro, instead of the micro. Why? As Kim Bowes, a professor of classical archaeology at the University of Pennsylvania, explains in Surviving Rome: The Economic Lives of the Ninety Percent, until recently the ordinary lives of ordinary Romans eluded us for lack of evidence. Only in the last three or four decades, thanks to an explosion of archeological digs often triggered by construction projects across Europe, have we been showered with new knowledge about the lives of what Bowes labels “the 90 percent.”  

“It’s a delicious irony,” she writes, “that more information about the rural Roman 90 percent has emerged from the construction of Euro Disney [about 15 miles east of Paris] than from the well-intentioned excavations designed to find them.”

Perhaps we assumed that “everyday working people” in ancient Rome didn’t write much about themselves. Certainly, the well-known chroniclers of the day — Sallust, Livy and Tacitus — didn’t focus on them; they mostly wrote instead about the big names who wielded political power. But thanks to discoveries of the past four decades — including graffiti, and writings on broken pottery and wooden tablets, coins and documents and farm implements, and scientific analysis of soil samples and ancient ruins — we’ve learned more about the lives of ordinary people in the Empire than historians ever knew before. 

This “shower of information about Roman farms and fields, crops and herds, and the geology and soil science,” Bowes argues, is transforming our understanding of life at the time. Her book is the first notable effort to tell the world what these recent findings reveal. 

Let’s remember that the history of ancient Rome did not begin with the Empire. For 500 years before its first emperor, it was a remarkable res publica (a republic) known for the rule of law, substantial liberty, and the dispersion of power. When that crumbled into imperial autocracy late in the first century BCE, what we know as “the Empire” took root and lasted another 500 years. Weakened internally by its own welfare-warfare state, the Western Roman Empire centered in Rome fell to barbarians in 476 CE. Bowes’ attention is drawn exclusively to that second half-millennia.  

The Empire evolved into a very different place from the old Republic. By the dawn of the second century CE, it would have been unrecognizable to Roman citizens of the second century BCE Loyalty to the state and one-man rule had largely superseded the old republican virtues. Many emperors were ghastly megalomaniacs to whom earlier Romans would never have groveled.

Despite the general decline in morals and governance that characterized much of the imperial era, ordinary Romans fared better than you might surmise, at least until the decline overwhelmed them in the late fifth century. Bowes attributes this to “cagey managers of small resources” who lived in “precarity” but “doubled down on opportunities.” The picture she paints with recent evidence is one of hard-working, resourceful farmers, tradesmen, and shopkeepers making the very best of a tough situation and, for the most part, doing remarkably well at it. 

Ongoing excavations at Pompeii, destroyed by a volcanic eruption in 79 CE, have yielded fascinating details of commerce and coinage in the city: 

Bar and shop owners did more of their business in bronze and less with the prestige metals. Resellers of bulk oil and wine, and above all artisans producing for larger-scale markets, used gold and particularly silver. 

Bowes reveals that much of the Roman world experienced a “consumer revolution” as the Empire stretched from the border with Scotland to the Levant and across north Africa. Roman roads and trade, even while government grew more tyrannical in Rome, facilitated it. Consider this finding:

New data from archeology, and newly reconsidered texts like the Pompeii graffito, find working people, even some of the poorest, consuming far in excess of our previous expectations. From small-time traders to enslaved servants, farmers to craftsmen, Romans ate more and different foods, purchased rather than made many of the items they used…Their levels of household consumption were thus historically quite high…

The immense quantity of data gleaned from the recent discoveries shows up in numerous tables, charts, graphs and illustrations in Bowes’ book. From those entries, we learn of the accounts of a Roman beer-buyer; the percentage of farms with lamps, candlesticks and window glass; which cereal crops were grown on small farms; the real incomes of artisans and shopkeepers; the prevalence of metabolic disease among children, and so much more.  

For comedians who often poke fun today at British teeth, there’s this tidbit: Data suggest that the Roman conquest of Britain brought dramatic declines in dental health and that “British urbanites had the same or perhaps even worse dental health as the mostly urban Italian sample.” 

Nonetheless, new evidence suggests that “the majority of Romans were consuming a relatively robust caloric package.” Bowes tells us, 

This meant a lot more energy to do work, and thus a lot more work could be done. The Coliseum was not built on 1,900 calories per day…The Coliseum was not built by workers scraping their porridge out of a single pot. 

So, we now know that ordinary Romans during the Empire likely lived better than historians previously believed. They exhibited “relentless persistence and shrewdness,” “perseverance and ingenuity,” a degree of “grit and hustle” we can appreciate more than ever. For several hundred years, their accomplishment “was their ability to wrest a living from a hard and complex world.”  

We also know that it didn’t last. As the Empire disintegrated in the fifth century, it became ever more difficult for many, and impossible for a great number, to eke out a living. The “Dark Ages” that commenced with the fall of Rome saw economic and cultural decline and a massive depopulation. Life spans shortened, mortality rose, and standards of living plummeted. At its height, the city of Rome itself was home to a million people; a few centuries later, it plunged to a nadir of barely 30,000.  

Though Bowes falls short of saying so herself, I think the moral of the story is this: A resourceful people can endure a great deal before they throw in the towel, but a thriving civilization depends on what the Roman Empire ultimately forfeited: peace, freedom, property rights, and the rule of law. 

One of the most robust findings in economics is that, with few exceptions, people respond to incentives, rather than intentions or moral principles. Individuals operate under constraints of time, information, and risk, and as such, they will predictably and understandably adjust their behavior to whatever metrics ensure success. To do otherwise is irrational. When performance is evaluated and rewarded using metrics like quotas, behavior shifts toward satisfying those quotas to secure the benefits thereof. This happens in firms, schools, hospitals, police departments, and regulatory agencies, even when everyone understands, at least in the abstract, that the metric is distinct from the goals to be achieved.

Immigration enforcement provides a vivid case study of this general institutional failure mode. Under recent policy changes, US Immigration and Customs Enforcement has operated under explicit arrest targets in the form of daily and annual numerical goals meant to demonstrate enforcement intensity and resolve. The political rationale for these targets is straightforward. It is to signal to voters and political supporters that the current administration is serious about protecting the border and clamping down on illegal immigration.

But economics teaches that what gets measured gets optimized and gamed for various reasons, mostly having to do with incentives. In the case of immigration enforcement, when success is defined in numerical terms, agents will pursue the cheapest path to those numbers, rather than pursuing individuals and groups that are harder to find and detain. That is a rational given the incentives created by the Administration, namely rewarding aggressive arrest quotas. It makes sense that whenever institutions or individuals face quotas, they are likely to focus on the low-hanging fruit. Time spent achieving an easy unit of output eats up time spent pursuing a hard one. Effort that is devoted to high-risk targets, like violent criminals and well-entrenched gangs, threatens performance metrics in ways that low-risk targets do not. When failure to meet quotas carries professional consequences, agents will avoid activities that jeopardize the count, even if those activities are more closely aligned with the stated mission.

The logic is straightforward. Violent criminals, gang leaders, and professional smugglers are difficult to locate and expensive to apprehend, often relying on networks of other people to help them evade detection. Pursuing such criminal organizations requires investigations, coordination across jurisdictions, surveillance, and uncertain outcomes, making it easy for agents to come up empty-handed. By contrast, unauthorized immigrants who are otherwise law-abiding are comparatively easy to find. They have fixed residences, work regular jobs, and their children often attend the local school. Many are already interacting with the state through legal channels, including standard immigration check-ins.

When arrest quotas rise, then, it’s no surprise that arrests have accelerated disproportionately among those who are easiest to find and arrest rather than those who pose the greatest threat. Recent data confirm this pattern. Enforcement activity has surged, but the majority of arrests involve individuals without prior criminal convictions, a distribution consistent with quota-driven optimization rather than threat-based prioritization. And given the career and political incentives behind meeting those quotas, it is what we should expect. This behavior is rational given the incentives; it would be surprising if agents behaved otherwise.

There is a deeper problem here, though, that Hayek can help us diagnose. Quotas assume that central authorities know in advance how enforcement effort should be allocated across a vast and heterogeneous landscape. They assume that arrests are sufficiently homogeneous, such that merely counting them captures what matters. They assume that the marginal value of the next arrest is roughly constant across contexts. And they make these assumptions, often, without the salient local knowledge needed. 

Here the analogy to central planning becomes illuminating. Central planners, like those in Cuba or the former Soviet Union, fail because they lack access to the dispersed, tacit, and constantly changing knowledge required to allocate resources efficiently. As Hayek argued, markets work not because anyone knows the right answer in advance, but because competition allows agents to discover it through decentralized experimentation and feedback information that would otherwise be unavailable. Enforcement environments share this complexity because, among other reasons, threats vary by region, network, industry, and time. A centralized quota cannot incorporate this information, partly because it treats arrests as interchangeable units in the same way that central plans treat tons of steel or bushels of grain as interchangeable.

This helps explain why quota-driven enforcement is insensitive to conditions on the ground. It cannot adapt to local threat profiles because it does not reward adaptation. It cannot prioritize effectively because prioritization is costly and quotas reward speed, and it cannot learn from failure because in most cases it lacks the local knowledge needed for the adjustment. Of course, politicians can pivot when citizens and voters push back, but it is necessarily a less detailed and efficient process than, for example, markets and prices. 

Worse still, enforcement that deliberately and disproportionately targets working, embedded individuals produces sudden and uneven labor supply shocks. Industries that rely heavily on immigrant labor, like construction and agriculture, experience disruptions that cascade via prices, output, and complementary employment. These are downstream consequences of enforcement choices shaped by quotas. When enforcement prioritizes ease of arrest over social cost, it predictably targets workers rather than criminals, disrupting productive relationships that markets had already coordinated. The result resembles what happens when planners disrupt supply chains without understanding their internal complementarities.

A common defense of quotas appeals to accountability. Without numerical targets, agencies may underperform, selectively enforce, or drift away from their mandates. That said, the existence of a real problem, namely accountability, is hardly a defense of a flawed solution based on quotas that measure a single dimension without the necessary local knowledge.

The central lesson is rooted in institutional design and incentive structures under which these immigration agents operate. When complex, knowledge-intensive activities are governed by centralized numerical targets, agents will rationally pursue targets in ways that undermine the broader purpose of the institutional effort. Perverse incentives and poor institutional design are not the only explanatory factors here —personal choice and moral character matter, too—but they are a big part of the explanatory pie.