For much of the last half-century, California benefited from a powerful first-mover advantage. Dense networks of talent, capital, and research institutions allowed the state to absorb policy mistakes that would have crippled competitors. High spending and taxes, restrictive housing rules, and regulatory complexity were treated as nuisances rather than binding constraints, because growth could outstrip their costs.

That margin of error has narrowed dramatically.

What California is now experiencing is not a cyclical tech downturn or a post-pandemic anomaly. It is a measurable, policy-driven decline in relative competitiveness. The most important evidence is not that tech employment has fallen in absolute terms, but that California’s share of national tech employment has been shrinking, while other states gain ground.

Markets are responding to incentives exactly as economic theory predicts.

Employment Share, Not Headlines, Tells the Story

According to Bureau of Labor Statistics Current Employment Statistics data, California’s technology employment growth has underperformed national trends for several years, including during periods when tech hiring stabilized or rebounded elsewhere, and recently has been declining. California’s share of US tech jobs is falling from roughly 19 percent pre-2020 to closer to 16 percent in recent years, a nontrivial shift for an industry this large.

This is a classic example of relative decline. California still employs more tech workers than any other state, but it is no longer where the marginal job is being created.

Commercial real estate data corroborate the employment figures. Office vacancy rates across Silicon Valley remain elevated well beyond what remote work alone would explain. Bay Area office markets have not recovered in the way peer regions have. Persistent vacancies signal not just a shift to hybrid work, but geographic reallocation of firms and labor.

Migration as a Labor Market Signal

Labor mobility reinforces the same conclusion. US Census state-to-state migration data show continued net domestic outmigration from California, particularly among working-age adults. While international immigration partially offsets population losses, domestic migration is more relevant for employer location decisions, especially in high-skill sectors.

Economic theory predicts that firms follow labor when relocation costs are low and regulatory frictions are high. California now faces both: high regulatory frictions at home and increasingly credible substitutes elsewhere.

Founding Versus Scaling: A Crucial Distinction

California still dominates early-stage venture capital totals, as shown in venture investment data. This is often cited as evidence that concerns about the state’s competitiveness are overstated. That interpretation conflates firm formation with firm expansion.

Founding activity reflects legacy advantages such as universities, networks, and capital concentration. Scaling decisions reflect marginal costs. Increasingly, firms are choosing to incorporate or raise seed funding in California while expanding headcount in lower-cost, lower-regulation states.

From an economic standpoint, this is predictable. Scaling in California exposes firms to the nation’s highest marginal income tax rates, comparatively punitive capital gains taxation, rigid labor mandates, slow permitting processes, and volatile regulatory expectations. These costs rise nonlinearly as firms grow.

AI Regulation as a Binding Constraint

Artificial intelligence policy may become the clearest illustration of California’s regulatory overreach.

A recent CalMatters analysis documents how California lawmakers have pursued some of the most expansive state-level AI regulations in the country. These proposals extend liability, mandate preemptive risk assessments, and impose compliance obligations before alleged harms are empirically demonstrated or even defined.

From an economic perspective, this approach treats innovation as a presumptive externality rather than a productivity-enhancing input.

AI is widely understood as a general-purpose technology. Research shows that such technologies generate broad, economy-wide productivity gains, not sector-specific benefits. Overregulating AI therefore depresses expected returns not only in software, but across healthcare, logistics, manufacturing, finance, and education.

California’s AI regulatory framework has drawn federal scrutiny, which is instructive. As noted in CalMatters, state-level AI mandates were referenced in Trump’s recent presidential executive order, citing concerns over fragmented and inconsistent state regulation. Regardless of political framing, the economic concern is straightforward: regulatory fragmentation raises fixed costs and discourages upscaling.

Regulation, Market Structure, and Incumbency

California’s regulatory posture also has implications for market structure. Extensive empirical literature shows that high fixed compliance costs reduce entry and increase concentration. The OECD’s work on regulation and competition consistently finds that heavier regulatory burdens favor large incumbents at the expense of startups and challengers.

This dynamic undermines the very competition that drives innovation. Europe’s experience with digital (over)regulation offers a cautionary parallel, acknowledged even in European Commission competitiveness reports. California risks reproducing that outcome domestically, exporting innovation to other states rather than other continents.

Costs Complete the Incentive Structure

AI regulation is best understood as the marginal constraint layered atop an already expensive environment. California has the highest top marginal income tax rate in the United States, and taxes capital gains as income. Housing scarcity, documented extensively by UC Berkeley’s Terner Center, raises labor costs without increasing real purchasing power. Energy prices remain among the nation’s highest, as shown by EIA electricity price data.

In combination, these policies alter the expected return on investment at the margin. States like Texas and Florida offer credible alternatives: no personal income tax, faster permitting, lower housing costs, and a lighter regulatory touch. 

Firms do not need ideological motivation to relocate. The incentive structure does the work.

Opportunity Costs and Distributional Effects

The economic cost of tech job relocation extends beyond headline employment figures. When tech employment relocates, these spillovers disappear as well. The distributional consequences are regressive. High-skill workers are mobile. Lower-income workers tied to local economies are much less so. Policies that suppress growth (even under the banner of equity) often hurt the poor most.

A Predictable Outcome

Unless California changes course, the trajectory is clear. AI firms will incorporate elsewhere. Venture capital will follow labor. Scaling will increasingly occur in states that treat innovation as an asset rather than a liability.

California will remain an important source of ideas. It will be a diminishing source of jobs. Markets are not ideological. They respond to incentives. On that front, the verdict is already in.

Recently on Facebook, I shared my Café Hayek post titled “Lower-Priced Goods are a Blessing, Not a Curse.” I prefaced this share with this remark: “Protectionism is the theory that 10+2=6, and 10-2=16. And protectionists proudly and tirelessly defend this theory, happy to dismiss as ‘elitists,’ ‘experts,’ or ‘globalists’ those of us who point out that 10+2=12, and that 10-2=8.”

Of course, my description of protectionism isn’t literally true. And yet it does truly capture protectionism’s essence, which is the bizarre belief that a greater abundance of goods and services made available from sources outside of a nation’s boundaries reduces the supplies of goods and services available to the people of that nation, while policies that diminish the abundance of goods and services made available from sources outside of a nation’s boundaries increase the supplies of goods and services available to the people of that nation.

Putting aside the national-security exception to the case for free trade, such an arithmetical impossibility is indeed what 90 percent of protectionism is revealed to be, when stripped of the vague and misleading language typically deployed to mask its essence. Tariffs and other protectionist interventions are sold as means of creating more and higher-paying jobs (which would, in turn, reverse the allegedly rising “cost of thriving”), of paving paths for the development of “the industries of the future,” of raising impressive amounts of tax revenue from foreigners, of making the economy more ‘competitive,’ and generally of strengthening the domestic economy, improving the living standards of ordinary citizens.

Because voters overwhelmingly like policies that promise them greater access to goods and services, protectionists understandably trumpet the alleged ability of protectionism to deliver on this economic front.

But what about the remaining 10 percent of the attempted justifications of protectionism (again, putting aside considerations of national security)? These justifications pretend to be non-materialistic and, therefore, presumably are ‘higher’ and more weighty than are ‘merely economic’ concerns. Such is the stance, for example, of Mr. Kang Chen, who offered this comment in response to my Facebook post: “No. Protectionism is the theory that there are things that matter besides the prices of goods and services.”

An easy response to a comment such as Mr. Chen’s – a response that’s accurate and appropriate despite its easiness – is to point out that the great majority of the pleas for protectionism promise improved material well-being. More jobs. Higher wages. Rising standards of living. A larger share of our tax revenues paid by foreigners. Protectionists such as Mr. Chen would be taken more seriously if the likes of Donald Trump and Elizabeth Warren campaigned for higher tariffs by explicitly announcing that “tariffs will significantly raise, now and into the future, the prices of the goods and services that you and all American households regularly purchase. Most of you, therefore, will see your real wages fall and your material standard of living worsen. But don’t worry! Your lower standard of living will be more than offset by non-material benefits.”

Protectionist politicians never say such a thing. On rare occasions, protectionists triumphantly declare that higher tariffs might reduce people’s access to cartoonishly frivolous luxuries such as “plastic baubles and trinkets,” or, as President Trump said last year, “maybe the children will have two dolls instead of thirty.” (Such declarations are meant to convince voters that the economic costs of protectionism are trivial compared to its economic benefits.) But never do protectionist politicians campaign on a platform of arranging for people to be economically poorer as a price to be paid for non-economic benefits.

No more need be said to dismiss Mr. Chen’s suggestion that protectionism, in practice, is about the sacrifice of economic well-being for higher non-economic ends. But more can – and should – be said.

What people such as Mr. Chen and others believe themselves to be doing when they insist that protectionism is about more than “the prices of goods and services” is distinguishing themselves from free traders, who are assumed to be concerned only with narrow material ends. Mr. Chen and Co. fancy themselves as standing in gallant opposition to the horde of mindlessly materialistic free-traders in order to promote ends such as job security, the family, and the character of towns and regions.

But Mr. Chen and Co. deeply misunderstand the case for free trade. It is not a case for the elevation of shallow materialism over profoundly important non-economic ends.

First, very many (most?) free traders — including myself — support free trade ultimately because it’s consistent with individual liberty, while protectionism is an offense against individual liberty. Even if free trade somehow resulted in a reduced material standard of living, I and many other free traders would still champion it because of its non-economic virtue of being consistent with freedom. It’s fair for Mr. Chen and other protectionists not to esteem individual liberty as highly as do we free traders. It’s unfair, however, and mistaken for protectionists to accuse us free traders of valuing nothing higher than material enrichment.

Second, all motives for economic action ultimately are non-monetary (that is, they’re not about accumulating money for the sake of accumulating money). Some of these motives are material in a narrow sense and, hence, might be called “materialistic”: everyone must eat and be housed and clothed. And some of these materialistic motives are indeed crass and shallow and even contemptible: Joe uses some of his monetary earnings to get drunk on Friday nights while Janet regularly feeds some of her monetary earnings into slot machines. But other of these motives are not materialistic in any narrow sense: Jane spends much of her monetary earnings on piano lessons for her grandchildren while Jerry donates a portion of his monetary earnings to a community children’s theater and uses another portion to improve his and his wife’s learning by subscribing to The Rest Is History podcast. Because free trade increases the opportunities to do all of these things, it’s erroneous to suggest that the case for free trade is a case only for narrow material or sensual gratification.

Third, nearly all of the alleged non-materialistic benefits of protectionism are, in fact, materialistic benefits.

Consider, for example, job security. Job security is valued largely because a secure job is a secure stream of income. If job security really were a non-economic goal that trumps ‘mere’ material well-being, workers who have this goal could greatly increase the security of their jobs by offering to take a significant cut in their monetary wages. Yet, tellingly, such wage-cut offers seldom occur. The case for using protectionism to increase the job security of workers in protected industries is the case for having fellow citizens other than the protected workers pay the economic cost of making the protected jobs more secure.

It’s admirable to have non-economic goals. But it’s detestable to force other people to subsidize the achievement of these goals, and hypocritical to accuse those of us who object to such subsidization of being excessively materialistic. If any group in this situation is excessively materialistic, it’s the protected workers and the protectionists who apologize for them. These protectionists never pause to ponder what non-economic goals a policy of protectionism prevents the bulk of their fellow citizens from pursuing. As a result of having to pay prices driven higher by tariffs, how much leisure does a working mom lose? How much does a family’s education budget shrink? How much health care must another family forego? By how many years does dad postpone retirement?

If protectionists are in search of people who are mindlessly and narrowly materialistic – of people who are blind to the non-economic goals that most individuals have – protectionists should look in the mirror.

Private markets work well when they are allowed to function free from big government directives. The latest example is in the world of financial data sharing, where Washington tried to dictate prices, warring industry interests warned of disaster, and the free market quietly delivered a solution on its own.

Since the passage of major financial industry reform legislation in 2010, policymakers have debated how to create a regulatory framework within which personal data portability and sharing can occur safely and securely. Unsurprisingly, the Biden Administration sought to finalize rulemaking that was supposed to put an end to some 15 years of speculation on what this framework, under Section 1033 of Dodd-Frank, would look like.

For years, banks were required to provide financial technology companies with customer data for free, with few limits on how that data could be monetized. The Biden Administration embraced that policy, finalizing a rule that required banks to share data with aggregators at no cost. It was a government-mandated price control of zero dollars, delivering a “solution” to a problem that didn’t exist.

President Trump saw the absurdity of the so-called Rule 1033 and acted swiftly to stop it. By forbidding banks from charging for access, the rule would have forced them to absorb the infrastructure, cybersecurity, and compliance costs of data sharing while allowing fintech firms — favored by the Biden Administration — to profit from the very data they received at no charge. The result would have been a system where the costs were socialized, and the profits were privatized.

After a federal court issued an injunction halting enforcement of the Biden-era rule, some claimed that government needed to put its thumb on the scale to negotiate fair compensation for access. Those arguments have already collapsed.

In recent weeks, JPMorgan Chase — the country’s largest bank — reached market-based pricing agreements with 95 percent of the data middlemen in the market. One such actor supporting greater government involvement went so far as to confirm that its updated agreement with the bank “wouldn’t affect current customer deals or pricing.” 

The success of these agreements makes one thing clear: market participants can solve complex problems through negotiation and partnership far better than regulators can through compulsion. Voluntary, market-based arrangements have succeeded where central planning failed.

As the Consumer Financial Protection Bureau prepares to issue an updated rule, it should reject the false premise that consumer privacy and innovation can only thrive under government control. Embracing a market-based framework will strengthen consumer protection and foster responsible innovation far more effectively than price mandates ever could.

The best course of action would be to eliminate the agency entirely, and transfer its powers to other areas in the federal bureaucracy. If that is not possible, there is a menu of options to reform the structure to better serve taxpayers. Over the past decade, the agency’s priorities have swung dramatically with each administration, creating instability for consumers and industry alike. Concentrating so much authority in a single director has proven unworkable. A bipartisan commission structure, like the Securities and Exchange Commission (SEC) or Federal Trade Commission (FTC), could restore accountability and consistency.

Congress should also bring the CFPB’s funding under the normal appropriations process, rather than letting it draw money directly from the Federal Reserve. Transparent funding would improve oversight and focus CFPB’s work on real consumer benefit instead of political goals.

Most importantly, CFPB should pursue modern oversight that protects consumers while supporting innovation. Its failure to evaluate the data security risks of mandatory data sharing under the Biden-era Rule 1033 shows the danger of heavy-handed policymaking.

President Trump’s CFPB should build a lasting legacy of practical, forward-looking regulation that respects both consumer protection and market competition. The fall of the Biden-era Rule 1033 is a reminder that the free market, when allowed to work, delivers the best results for everyone.

When the internet went mainstream at the turn of the twenty-first century, it was widely celebrated as a revolutionary force for freedom and democracy. Its decentralized architecture promised to empower individuals, expand free expression, and weaken the grip of authoritarian states. Many believed that open information flows would make censorship obsolete and repression impossible to maintain.

That optimism has not merely faded — it has been decisively overturned. The same technologies once hailed as instruments of liberation are now being repurposed as tools of surveillance, censorship, and control. What is unfolding is not a sudden collapse of digital freedom, but a slow, structural transformation of the internet itself — one that is quietly reshaping how power operates in the digital age.

Crucially, this shift is not confined to authoritarian regimes. It is increasingly spreading into liberal democracies that once saw themselves as custodians of an open and global internet.

A Global Recession of Digital Freedom

Internet freedom is deteriorating globally at an unprecedented rate. The Freedom on the Net 2025 report marks the fifteenth consecutive year of decline, representing the longest recorded recession in digital freedom. Nearly 80 percent of internet users live in countries where a social-media post could result in arrest, and two-thirds are in nations where people have been assaulted or killed for their online expression. Governments in 65 percent of assessed countries block political, social, or religious content, while more than half restrict access to major platforms altogether.

This erosion is no longer confined to the usual suspects. Even established democracies are backsliding. U.S. internet freedom fell to a record low in May 2025, dropping three points in a single year — reflecting a growing willingness among democratic governments to deploy tools once associated with authoritarian rule. 

That shift is already visible in practice. During the unrest in New Caledonia, France restricted access to TikTok; meanwhile, authorities in the United States, India, and Brazil have pressured platforms to remove political content. At the same time, Meta and X have rolled back transparency tools that once enabled researchers to track disinformation and government influence.

Two decades ago, such a trajectory would have seemed implausible. In 2000, President Bill Clinton famously mocked China’s early censorship efforts, likening them to “trying to nail Jell-O to the wall.” Yet China went on to build the Great Firewall — the most comprehensive censorship system in modern history — reshaping global assumptions about what information control could achieve. 

What once appeared to be a uniquely authoritarian experiment has since evolved into a widely adopted model of digital governance, replacing the open, participatory internet imagined in the 1990s with a controlled, increasingly surveilled digital environment shaped not by a single censor but by the combined pressures of regulation, corporate incentives, and algorithmic control.

How the West Is Quietly Adopting Authoritarian Tools

In Western democracies, digital control rarely takes the form of overt repression. It advances quietly — through regulatory creep, technical adjustments, and procedural changes that seldom provoke public alarm. Surveillance expands, encrypted spaces shrink, and the line between state authority and corporate power blurs. Control is not imposed as repression but framed as protection, administered through law, normalized by bureaucracy, and legitimized by democratic institutions.

Encrypted communication — once indispensable for journalists, activists, and dissidents — is increasingly under threat. Europe’s proposed Chat Control legislation would require scanning private messages, weakening end-to-end encryption by mandating content inspection before or after transmission. In parallel, the UK’s Online Safety Act and Australia’s identity-verification rules introduce new points of access to private communication — often justified in the language of safety or child protection.

If eroding encryption compromises private communication, mandatory digital identity systems go further by undermining anonymity itself. Proposals such as the UK’s BritCard, the EU’s Digital Identity Wallet, and similar frameworks in Australia and parts of the United States would link online activity to state-verified identities. When integrated with corporate datasets — biometric, location, financial, and browsing data — these systems enable continuous monitoring without requiring explicit surveillance orders.

Much of this infrastructure is now supplied by private firms rather than built by the state. Companies such as Palantir have become central actors, providing data-fusion platforms to intelligence agencies, police forces, militaries, and immigration authorities across multiple (otherwise) democratic states. What began as narrowly framed security tools has evolved into systems that aggregate vast troves of personal data and deploy predictive analytics at scale — raising profound concerns about bias, accountability, and oversight, even in societies long committed to strong privacy protections.

Together, these tools are beginning to form an integrated surveillance system. Digital identity systems, predictive algorithms, and surveillance technologies increasingly reinforce one another, creating a state–corporate surveillance architecture that — though softer and more bureaucratic — mirrors key features of digital authoritarianism. Platforms police users to meet regulatory demands, while governments rely on private firms to enforce political priorities. Control expands not through overt repression but through routine administrative processes that quietly shrink the space for individual freedom.

Pavel Durov’s warning is therefore not an exaggeration. Digital liberty is not taken away all at once. It erodes through the accumulation of quiet legislation, routine deployments, and the gradual weakening of institutional safeguards, which slowly remake the internet into something it was never meant to be. If freedom is to endure, it must remain the starting point of governance — not its exception.

Otherwise, Durov’s warning may be remembered not as a call to action, but as a record of freedoms already gone.

The US seizure of Venezuelan leader Nicolás Maduro is being framed publicly as a counternarcotics and democracy-restoration operation. But it is oil — not cocaine or fentanyl — that sits at the center of events. Venezuela’s vast reserves, its role in gray and black energy markets, and its position within a broader geopolitical contest over oil supply explain far more about the timing and scope of the intervention than narcotics enforcement ever could.

Venezuela is no longer the oil superpower it once was. Production has collapsed from more than three million barrels per day in the late 1990s to under one million today, placing the country outside the top tier of global producers. Still, oil remains the backbone of the Venezuelan economy, accounting for roughly 95 percent of export revenue. In a world where energy markets are increasingly shaped by sanctions, supply fragmentation, and political risk, even marginal barrels matter — especially when they are sold at a discount and routed outside formal channels.

In recent years, Venezuelan oil has flowed largely into opaque markets, particularly to China, often via intermediaries and “ghost ships” that mask origins to evade sanctions. These barrels are not priced at global benchmarks; they are sold cheaply, quietly, and strategically. The result is not simply lost revenue for Caracas, but distorted price signals across the global oil market. Interventions disrupt price discovery. Sanctions do not eliminate supply — they reroute it into less-transparent channels, where prices convey less information and capital allocation becomes more politicized.

The US blockade and seizure of sanctioned tankers and the disruption of naphtha imports, critical for transporting Venezuela’s heavy crude, had already begun constraining production even before the military operation. Storage tanks filled, wells were shut, and exports stalled. Yet global oil prices barely moved. That muted response reflects a market already awash with supply and conditioned to treat Venezuelan output as unreliable. Oil markets have learned to discount politically fragile production, which means that sudden interventions often have less immediate price impact than policymakers expect.

The longer-term implications, however, are more significant. A successful political transition followed by large-scale foreign investment could eventually bring Venezuelan production back toward its pre-collapse levels — perhaps to 2.5 million barrels per day over several years. That would represent a meaningful supply shock, potentially lowering global oil prices by several percentage points over time. Such an outcome would benefit refiners, particularly in the US, that are configured for heavy crude, while putting downward pressure on higher-cost producers elsewhere.

But that optimistic scenario rests on fragile assumptions. Oil production is not simply a matter of drilling holes; it requires institutional stability, secure property rights, skilled labor, functioning infrastructure, and credible contracts. Venezuela’s oil collapse was not caused by geology, but by decades of state control, politicized management, expropriation, and capital flight. Reversing that damage will take time and discipline.

There is also a broader pattern worth noting. Within a single week, the United States has been exerting escalating pressure on three oil-producing nations across three continents: Venezuela, Iran, and Nigeria. Whatever the specific justifications in each case, the pattern suggests a strategic shift. A decade ago, Donald J. Trump rose to prominence as an anti-interventionist critic of foreign entanglements. Today, the US is asserting itself as an active enforcer of energy order, using sanctions, seizures, and force to reshape supply flows.

Among other reasons, it matters because energy markets thrive on decentralized discovery and suffer under centralized control. When oil becomes an explicit instrument of geopolitical maneuvers, prices reflect power as much as scarcity. Capital flows follow political signals rather than entrepreneurial ones. The result is not necessarily higher prices, but noisier ones: prices that convey less reliable information about underlying supply and demand.

Discounted oil sold into black markets sustains regimes, finances patronage networks, and reshapes global trade patterns. Controlling that flow is economically consequential in a way that narcotics interdiction rarely is. Whether the US intervention ultimately stabilizes Venezuela or entrenches a prolonged foreign presence, its lasting impact will be felt less in Caracas politics than in the structure — and credibility — of global oil markets.

W.E.B. Du Bois was born in Great Barrington, Massachusetts (where AIER is now headquartered), in 1868. Today, this towering figure of the early civil rights movement is remembered as a groundbreaking sociologist, Pan-African socialist, and near-mythical hero to the intellectual left.

“He’s a reformist,” philosopher Cornel West told a classroom of Dartmouth students in a 2017 lecture on Du Bois’ long path to becoming a revolutionary. “But he’s a radical reformist, no doubt.”

But there was once a W.E.B. Du Bois who was radical mainly in the scientific sense. Before drifting into the study of history and sociology, he was an economics student at Harvard. The marginal revolution had just remade the dismal science into a more mathematical and literally “edgy” subject. And Du Bois made original contributions that leveraged insights from the free-market Austrian school and anticipated later developments in neoclassical economic thought, as Daniel Kuehn explains in a recent paper published in the Journal of Economic Perspectives.

Similarly, the young Du Bois’ recommendations for black racial uplift bore surprising similarities to the modern-day conservative economist Thomas Sowell. What caused his later radicalization? It was arguably a tragedy of racism.

Du Bois’ maternal great-great-grandfather was born in Africa and enslaved in America. But in the late 1700s he gained his freedom, possibly by fighting in the American Revolution. By the time Du Bois was born in Great Barrington, the town had a small but largely integrated black population. Du Bois’ mother (his father had abandoned the family when Du Bois was a toddler) owned land, and he learned and played at the public school alongside white kids. In 1888, having already studied at historically black Fisk University, he became only the sixth African American student to matriculate at Harvard.

Studying under Frank Taussig, Du Bois wrote a 158-page essay titled A Constructive Critique of Wage Theory. It included a thorough review of Carl Menger, one of the drivers of the marginal revolution, and his insight that the market value of goods and services does not depend on the value of inputs, but rather the value that consumers place on the most recent, or marginal, unit of output.

In his essay, Du Bois built on such work and rigorously demonstrated what Kuehn terms “a statement of wages as equal to the marginal revenue product… Du Bois identifies this need to think in terms of what would ultimately be called the marginal revenue product of labor.”

Kuehn goes on to note that Du Bois provides “one of the earliest acknowledgements that a labor-leisure trade-off determines individual labor supply in the marginalist framework.”

A year later, Du Bois left Harvard for two years of study at what is today the Humboldt University of Berlin. There he was exposed to a more historical approach to economics under scholars such as Adolph Wagner. Du Bois’ interests evolved, and when he returned to Harvard to finish a PhD (the first PhD Harvard would award to an African American), it was in history.

In his autobiography published in 1968, Du Bois would look back and characterize the economics he studied under Taussig as “reactionary” and “dying.” But as a newly minted PhD, Du Bois still had a long way to go to reach that point. His early works such as The Study of the Negro Problems (1898), The Philadelphia Negro (1899), and The Negro in Business (1899, which he edited), mention family cohesion, productive skills acquisition, and entrepreneurship as keys to black uplift. The required precursor, he believed, was ending racial discrimination.

But having taken a position at Atlanta University in Georgia, Du Bois was immersed in the South’s era of Jim Crow segregation. It was a time when a black man accused of a heinous crime against whites could find himself facing, rather than a court of law, mob action determined to surpass in barbarity the alleged underlying crime. Sam Hose was such a man, alleged to have murdered his white employer in 1899. A mob kidnapped him from a jail in Newnan, Georgia, dismembered him and burned him alive. Another black man was shot to death for “talking too much” about the attack on Hose.

Du Bois later reported in his autobiography that on his way to meet an Atlanta newspaper editor to discuss the lynching, he learned the burnt knuckles of Hose’s hand were on display in a nearby store window. He said the experience “broke in upon my work and eventually disrupted it…one could not be a calm, cool, and detached scientist while Negroes were lynched.”

Was this the final disappearance of the W.E.B. Du Bois who had once made those economic breakthroughs at Harvard? Subsequent years saw him drift to the left. In 1910, Du Bois joined the Socialist Party of America. In 1926, he visited the new Soviet Union, which he saw as a beacon of hope for racial equality. In 1961, he joined the Communist Party USA. By this time, he seemed to believe that, rather than having potential for black uplift, capitalism was an obstacle to it.

The suffering of The Great Depression likely played a role in his views, as it did for some others. But one wonders how much Du Bois’ embrace of socialism had to do with the simple fact that, for all their proven faults, such regimes tend not to be concerned with skin color. They oppress all races the same. 

We live in a time when many young people have a similarly friendly view of socialism. They see the historic wealth produced by free markets not as a path to their dreams but an obstacle to them. And like the evolution of Du Bois’ economic thought, it’s a tragedy.  

Bitcoin and other cryptocurrencies are widely – but wrongly – panned as unregulated casinos or Ponzi schemes that create no real value. For example, US Senator Elizabeth Warren called crypto a “threat to financial stability,” while the UK’s Treasury Select Committee said that cryptocurrency ownership “more closely resembles gambling than a financial service.”

While some cryptocurrencies are mainly speculative, many serve specific business or functional purposes. We can identify some of the value created by cryptocurrencies by breaking them into four general categories: Bitcoin, stablecoins, meme coins, and utility tokens.

1. Bitcoin

Bitcoin (BTC) is the original cryptocurrency. It is the base token of the Bitcoin protocol, a decentralized proof-of-work blockchain based on the 2008 whitepaper by Bitcoin’s anonymous founder Satoshi Nakamoto. The protocol has a limited supply, with an eventual maximum of 21 million bitcoins.

Unlike most cryptocurrencies, Bitcoin has only one purpose: to be used as money – or, more specifically, as a system of payment. It has no other features. The Bitcoin network is decentralized, which makes it highly resilient and hard to disrupt, though coin prices can be quite volatile.

With a market capitalization of around $2 trillion, Bitcoin is by far the largest cryptocurrency by market value. No other blockchain has anywhere near its history, its reliability, or its dedicated flock of fans and users. Bitcoiners often say that “Bitcoin is not crypto” because it is so fundamentally different from other blockchains that it deserves a category of its own.

2. Stablecoins

Stablecoins are tokens whose value is tied to a particular asset, most commonly the US dollar. They are widely used in electronic payments since they provide the benefits of blockchain-based payments without Bitcoin’s price volatility. Stablecoin payments are especially prominent in countries with unstable national currencies, whose governments cannot be trusted to maintain the value of their money.

The two most widely used stablecoins, Tether (USDT) and Circle’s USDC, have market capitalizations of about $148 billion and $62 billion, respectively. Both tokens are readily redeemable for US dollars. Circle is regulated as a money transmitter in the United States. Tether is a foreign entity, but is in the process of launching a regulated US subsidiary.

The opposite of gambling, stablecoins are safe, stable assets that serve as an electronic version of US dollars.

3. Utility tokens

Utility tokens are cryptocurrencies created by blockchains that provide some utility or service.

One example is Filecoin (FIL), which offers online storage, likeiCloud or Microsoft OneDrive, but on a decentralized public blockchain. The Filecoin blockchain provides safe and private file storage on the decentralized Filecoin network. The FIL token is used to pay for storage and is paid to participants to provide storage space on the Filecoin network.

A subset of utility tokens known as Decentralized Physical Infrastructure (DePIN) uses decentralized blockchains as a replacement for government or corporate-based infrastructure.  The Helium network (HNT), for example, provides a block-chain based marketplace for buying, selling, and transmitting WiFi and mobile phone data.

In addition, the decentralized finance (DeFi) industry is building a parallel financial system on blockchain technology, which is cheaper and more transparent than traditional exchanges. Larry Fink, CEO of Blackrock, the world’s largest asset manager, has said that the tokenization of traditional assets will be “the next major evolution in market infrastructure.”

Unsurprisingly, utility tokens – those with actual functionality and business purposes – tend to be the category most attractive to major cryptocurrency investment funds and venture capitalists.

4. Meme coins

There is one category of crypto tokens meant purely for speculation: Meme coins. These tokens have no functional purpose and no intrinsic value aside from the fun of trading. They are based on “meme” characteristics, like symbol or story that drives their prices. Many use pictures of dogs, frogs, and hats. The most popular meme coin DOGE, represented by a picture of a Shiba Inu dog and frequently referenced by Elon Musk, has a market cap above $26 billion. Another token, FARTCOIN, is based on, well, fart jokes.

There are political meme coins for candidates like BODEN and TREMP, whose prices bounced around before the 2024 elections as candidates moved in and out of favor, with both eventually crashing. After the election but before taking office, President Trump launched his own meme coin TRUMP, which peaked in late January, then lost 80 percent of its value within a few months.

Most meme coins trade for the fun of participating in a shared joke or the excitement of betting that the price will rise. They are indeed gambling in the truest sense, but despite these tokens being amongst the most well-known to non-crypto folk, this category of tokens represents only a small segment of the crypto market.

While there is certainly much speculation in cryptocurrencies, as in all financial markets, the cryptocurrency industry is more than meme coins. Bitcoiners hope Bitcoin will become the world’s dominant means of payment or at least a common reserve currency. Stablecoins provide an efficient means of payment and a relatively stable store of value, at least as far as the US dollar does. Utility tokens create real value or serve some business function.

Collectively, cryptocurrencies provide a variety of functions and use cases, ranging from specific business purposes to no purpose at all. Users can gamble if they want to, or they can make more informed strategic investments. Crypto is more than just a meme coin casino.

Recently, two Federal Reserve governors delivered speeches with interesting differences. Michael Barr warned against weakening bank supervision, citing “growing pressures to scale back examiner coverage, to dilute ratings systems” that could lead to a crisis. Stephen Miran countered that “regulators went too far after the 2008 financial crisis, creating many rules that raised the cost of credit” and pushed activities into unregulated sectors.

Both governors make valid observations about their respective concerns. Yet neither addresses a more fundamental problem: the regulatory cycle itself may be the primary source of financial instability. Rather than preventing crises, financial regulation tends to shift risks to new areas, setting the stage for different—not fewer—failures.

The Regulatory Ratchet

Barr himself describes the pattern: “time and again, periods of relative financial calm have led to efforts to weaken regulation and supervision…often had dire consequences.” But this observation cuts both ways. Periods of crisis lead to regulatory overreach, which creates unintended consequences, which leads to calls for reform—and the cycle repeats.

The Savings and Loan crisis of the 1980s and early 1990s illustrates this dynamic clearly. Following widespread S&L failures, regulators imposed stricter capital requirements through the 1988 Basel Accord. Financial institutions responded by using securitization to reduce their regulatory capital requirements while maintaining risk exposure—creating the shadow banking system that would later amplify the 2008 crisis. The new regulations didn’t eliminate risk; they relocated it to where regulators couldn’t see it.

After 2008, the pattern repeated. Dodd-Frank increased capital requirements and restricted proprietary trading through the Volcker Rule. As Miran notes, “many traditional banking activities have migrated away from the regulated banking sector” because regulatory costs made these services unprofitable for banks. Credit migrated to private credit funds, collateralized loan obligations, and other non-bank lenders. 

Today, private credit markets exceed $1.5 trillion, largely outside regulatory oversight. When the next crisis arrives, it will likely originate in these sectors—not because markets failed, but because regulation distorted incentives and redirected risk to less efficient channels. “Shadow banking” now accounts for $250 trillion globally, nearly half of the world’s financial assets, with minimal regulatory oversight.

Managing Risk, Not Preventing It

This regulatory cycle reveals a deeper problem with how policymakers think about financial stability. Both prevention-focused regulation (Barr’s preference) and “peeling back regulations”
(Miran’s approach) assume regulators can outsmart markets. Neither addresses the knowledge problem at the heart of financial regulation: regulators are always fighting the last war while markets adapt faster than rules can be written.

A more effective approach recognizes that financial risk cannot be eliminated—it can only be managed when it materializes. Financial regulation, if there is going to be any, should focus on crisis resolution rather than crisis prevention. This means three things:

First, establish clear rules about who bears losses when failures occur. Uninsured creditors, not taxpayers, should absorb losses. The FDIC’s resolution authority works precisely because it allows banks to fail in an orderly way, with clear priorities for claims. Extending this principle—making “too big to fail” institutions write “living wills” that detail how they would be unwound—creates market discipline without micromanaging risk-taking.

Second, eliminate implicit guarantees that encourage excessive risk-taking. When creditors believe regulators will intervene to prevent losses, they stop monitoring risk carefully. The 2008 bailouts reinforced expectations of government support, which may explain why risk-taking continued despite stricter regulations. A credible commitment to let failures happen—even of large institutions—would do more to encourage prudent lending than any capital requirement.

Third, simplify the regulatory framework itself. Complex rules create opportunities for regulatory arbitrage and make it harder for market participants to understand their actual risk exposure. Miran identifies one such complexity: leverage ratios that penalize holding safe assets like Treasury securities, creating “contradictory incentives” that distort markets rather than stabilizing them.

Canada’s experience offers a useful contrast. Canadian banks weathered the 2008 crisis better than their American counterparts, despite having less stringent capital requirements and a more concentrated banking sector. The key difference? Canadian regulators focused on ensuring orderly resolution of failures rather than preventing all risk-taking. Banks faced real consequences for poor decisions, which encouraged more conservative behavior than any amount of supervision could mandate. Since 1840, the United States has experienced at least 12 systemic banking crises—Canada has had zero. During 2008, Canadian banks maintained an average leverage ratio of 18:1 compared to over 25:1 for many US banks. The US bailed out hundreds of banks; Canada bailed out zero.

Breaking the Cycle

The debate between Barr and Miran represents the latest turn in the regulatory cycle. Both assume their preferred approach will prevent the next crisis. History suggests otherwise. Until policymakers recognize that financial regulation shifts rather than eliminates risk, we will continue cycling between crisis, overreaction, unintended consequences, and the next crisis.

The alternative is clear bankruptcy procedures and eliminating implicit guarantees. Let markets—not regulators—price risk. Let banks—not bureaucrats—manage portfolios. And most importantly, let failures happen to those who take excessive risks, ensuring that profits and losses remain where they belong: with the institutions that make the decisions.