The smartest tax doesn’t have to raise a lot of revenue to do social good. The point of a carbon tax, for example, is to reduce greenhouse gas emissions to mitigate climate change. If it succeeds in incentivizing people to use less fossil-fuel-intensive energy, the economic gains would be enormous even if it raised little revenue.

When assessing the value of a tax, we should look at both its potential revenue effects and why it might “fail” to generate this potential revenue. If a revenue failure is driven or offset by the discouragement of socially undesirable outcomes (like pollution), then it still might be a smart tax.

This logic applies starkly to a financial transactions tax (FTT). An FTT is a tax that is a tiny percentage of the value of any financial transaction. An FTT with a base rate of 0.25 percent of each stock trade should be a priority the next time a progressive majority holds power in the United States. The FTT should apply to all financial transactions, not just stock trades, and the rate should vary by type of transaction but should be set to eliminate any possibility of tax arbitrage between different financial instruments.

How Much Could an FTT Raise?

Currently, the potential tax base for an FTT is astronomically large—the annual value of financial transactions is measured in quadrillions of dollars. But there are some sharp limits to an FTT’s revenue potential relative to this base: Because information technology and computerization have reduced the transaction costs of financial activity to just hundredths of a percent of the value of the underlying trade, even a very small FTT is large relative to these tiny underlying transaction costs. Given this, the tax drives a relatively large increase in transaction costs, and many of these trades would no longer be undertaken at this significantly higher cost. Essentially, an FTT would cause transaction costs to rise enough relative to the status quo to crowd out at least half, and maybe closer to 80 percent or more, of current financial transactions.

Still, even with this large amount of crowd-out, the estimated revenue of an FTT is significant. For example, in 2016, a co-author and I estimated that a 0.25 percent FTT on stocks could raise close to 0.75 percent of GDP, which today would be roughly $225 billion each year. That’s enough money to permanently extend the more generous child tax credit of 2021—a credit that nearly halved the rate of child poverty in this country—as well as to finance universal high-quality pre-kindergarten for every 3- and 4-year-old in the nation.

It would also be a very progressive tax—the incidence of the FTT would mirror the distribution of financial wealth in the U.S. economy (the richest are, after all, the people who make most financial transactions). The top 1 percent would bear about 40 percent of its burden, and the top 10 percent would bear over 90 percent.

Financial Gambling Isn’t Worth Protecting

But could it be precisely the limits to the FTT’s revenue potential that make it such a useful tax? If a 0.25 percent tax on a transaction renders that transaction no longer economically valuable to either party, is it really so bad if it doesn’t happen? Further, what if that transaction foisted costs on parties outside of the transaction (externalities, in the jargon of economists)? In this case, a tax that makes the transaction no longer profitable for anybody also leads to widespread social benefits. Carbon taxes, for example, impose additional costs for using energy generated by fossil fuels. They induce customers to switch to other forms of energy and should be expected to shrink the fossil fuel sector while growing the clean energy sector. However, even people who were not engaged in any transaction involving fossil fuels before the imposition of a carbon tax will benefit from this shift, as the harmful pollutants emitted by fossil fuel use will be reduced.

Are we sure financial transactions are like this—near worthless even to the parties involved in them on net and actively harmful to outsiders? Obviously not every financial transaction fits this bill. Lots of us have mortgages and auto loans, and we’re very happy to be able to borrow this money. And lots of us have 401(k)s that are invested heavily in corporate stocks, which go up and down in price every day. But the useful part of the finance sector is not the part that allows (let alone encourages) as much day-to-day trading as is technologically possible, but the part that allows the non-wealthy to mimic the wealthy in investing in diversified, broad-based funds that follow the steady upward trajectory of stock prices and dividends over time. This requires a robust financial sector, but not one that generates the sheer volume of trading that we see today. Arguing that lots of today’s hyper-volume in trading could be squeezed out by an FTT with no effect at all on anybody not directly in the finance sector is just recognizing that every day-to-day movement in stock prices is essentially zero-sum.

One example highlights how much financial churn there is in the economy relative to the underlying flows of goods and services that people need. In 2025, global trade—all global exports and imports—was estimated at $35 trillion. This trade obviously required a corresponding financial transaction of $35 trillion as well (dollars must be converted to euros for U.S. consumers to buy imported wine, for example).

But foreign exchange in financial markets is more than 100 times this large. Derivatives (which can include foreign exchange trades) are also 100 times as large. Surely there is room for some of these staggeringly numerous financial transactions to be phased out without harming a typical household’s access to imported goods.

The vast majority of these excess trades in financial markets are purely zero-sum—they are gambles on whether asset valuations will rise or fall in the very near term. Each person’s win in these bets is another person’s loss.

The Costs of Zero-Sum Finance

Even if all these zero-sum bets just canceled one another out and didn’t affect anybody outside of them, there would be good (if paternalistic) reasons to shrink their volume. Cryptocurrency trading, sports betting, and meme stock trading are series of zero-sum bets that see widespread losses among most participants while a privileged few (including those running the exchanges) make large profits at others’ expense. We could shrug and say that consenting adults should be allowed to regularly lose money, but this would constitute a pretty blithe underestimation of the effectiveness of modern advertising and the cognitive glitches that make us susceptible to it. And simply as a matter of fact, the evidence is mounting that participating in these kinds of financial markets is leading to economic distress for many.

But paternalism aside, the more compelling reason to shrink the volume of financial transactions is that they do not stay contained in a series of zero-sum bets that affect nobody else. Instead, these transactions are how the powerful finance sector makes its money. And how finance makes its money inflicts large costs on the rest of us, in two big ways.

For one, finance is prone to periodic crises that impose direct and indirect cleanup costs on the rest of society. The financial crisis that accompanied the bursting of the housing bubble after 2006 likely added materially to the depth and length of the following Great Recession. Even in narrow fiscal terms, it led to hundreds of billions of taxpayer dollars being dedicated to bailing out banks and other financial institutions. The fact that most of these taxpayer bailouts were eventually paid back does not mean that money was not at risk or that people should not have been angry that the financial sector went to the front of the line for help during that crisis. The other, even more important factor is that finance steadily and inexorably skims money every day from the rest of the economy that is not justified by any use value the industry provides. In short, we can have the valuable parts of finance—the payments processing, the maturity transformation, the matching of lenders and borrowers—at significantly lower cost.

FTTs Complement Smart Regulation

Finance regularly enters crises. It is shocking how short memories are about this. The Great Recession of 2008-09 and the ensuing lost decade of economic growth arguably shaped society and politics in the United States and in countries around the world. The financial sector’s role in recklessly providing tinder for the crisis led to a series of efforts to regulate it more tightly, including the Dodd-Frank Act of 2010 and the creation of the Consumer Financial Protection Bureau (CFPB).

And yet the CFPB has been effectively destroyed by the second Trump Administration, even after it managed to transfer billions of dollars from unscrupulous financial institutions to typical households. The provisions of Dodd-Frank have been steadily weakened over time. It took just 15 years to go from crisis to the dismantling of the institutions meant to avoid the next crisis.

As these institutions have been forcibly eroded, new sources of potential excess and crises in finance have emerged. The two most obvious are the rise of cryptocurrencies and the growing importance of private (and hence opaque) credit and equity markets.

These new forms of potential financial market instability should be regulated and supervised far more tightly than they currently are. Eventually they will likely cause some kind of crisis, and regulatory scrutiny will probably increase to some degree. But in the meantime, an FTT would be a huge help in this effort. Markets for cryptocurrency—which still has near-zero use value—are by definition nothing but zero-sum trades. Even a small FTT would likely crowd out significant numbers of crypto transactions. The smaller the market is, the easier it is to regulate—not least because the lucky few who have made fortunes at the expense of the many will no longer have the resources to cajole and bully politicians who look to undertake sensible regulation.

Putting a Stop to the Skimming

Finance was tightly regulated in the 30 years after World War II. Incomes in finance accounted for just 3.4 percent of GDP in those years. By 2006, the peak year of deregulated finance, they had more than doubled—accounting for 7.6 percent of GDP. The crisis of 2008 saw them fall to 5.1 percent, but they have since climbed back to 7.4 percent in 2024.

The extra 4 percent of GDP claimed by finance in 2024 relative to its 1948-79 average amounts to about $1.2 trillion. Notably, the share of employment accounted for by the finance sector has risen far more slowly, going from 2.6 percent to 4 percent between 1948 and 1970 and generally remaining between 4 percent and 5 percent thereafter. Only financial sector incomes, not workforce size or work hours, saw a huge increase in this period.

Higher productivity does not explain these income gains. Careful research has uncovered the stunning fact that finance has seen essentially no productivity growth—defined as the cost of intermediating a given value of assets—for decades.

What does help explain the higher incomes in finance since 1979 is the growth of actively managed assets. More of the economy’s savings is now managed by finance sector professionals who charge high fees for this service. This active management generally implies lots of churn of assets as one fund’s manager bets with another fund’s manager about which assets’ price will rise and which will fall.

But this active management, and the fees it generates, does not lead to higher average returns relative to just buying and holding index funds—funds that passively mimic the economy-wide portfolio of assets. Any extra return managers achieve barely covers (if at all) the costs they charge for this active management. For savers looking to earn the highest return on their assets, it is literally money spent for nothing.

For society at large, it’s a lot of highly educated workers in prime real estate undertaking jobs that generate income for themselves but do nothing to boost the wealth of savers or allocate capital more effectively. If those workers and that capital were put to more productive use doing things the economy genuinely needs, we could all be quite a bit better off.

Even worse, the huge incomes in finance distort other markets. For example, it is well known by now that U.S. health care is far more expensive than health care in other rich countries. One driver of this is higher salaries for physicians (particularly specialty physicians). A paper highlighting this fact makes an obvious point: “One explanation for the higher incomes of U.S. physicians may lie in the broader U.S. income structure. The share of income received by people in the top 1 percent of the U.S. income distribution far exceeds the corresponding share in the comparison countries.”

The intuition is simply that prospective doctors need to expect incomes large enough to keep them from pursuing high-salary careers in other sectors—most notably finance. Doctors are always going to be in the upper reaches of the income distribution, which seems appropriate as they are well-trained, accomplished people. But because the U.S. economy is dysfunctional in many ways—including a hyper-lucrative finance sector that does not deliver large social benefits—that drive vastly disproportionate growth in these upper reaches of the income distribution, the United States will of course also have to pay more for doctors. This point applies to doctors’ salaries and to many other aspects of the medical-industrial complex (pharmaceutical companies, device-makers) as well as a bunch of other industries.

Why is finance unique in its ability to skim so much unearned income from the rest of the economy? Lots of reasons. Market concentration in finance is extreme, and this stifles competition. Nobody can just start a mom-and-pop hedge fund to offer competition to existing financial institutions when they go wrong. Further, principal/agent problems are also extreme in finance. Very few of us invest directly—we entrust our money, for good reasons, to professionals (the agents to our principals). It’s true that day traders exist, and in theory you could manage the investment of savings yourself using tools like Robinhood, but most of us have day jobs (and a healthy fear of failure) that make doing this impractical. The problem with these principal/agent relationships is that money managers are looking to maximize their own income, not our returns, and regulation is not strong enough anymore to keep those two goals well aligned. Additionally, asymmetric information means that money managers know much more than their clients about financial markets and how to make investments. We can look at broad benchmarks like the S&P 500 and will eventually notice if our money manager starts severely underperforming them, but the question “Could I have gotten the return I got with lower fees?” is very hard to answer from the outside.

Market failures exist outside of finance. What is so maddening about finance, however, is that these failures and the concomitant need to regulate and supervise financial institutions were recognized for decades until the 1980s. Once we unleashed the genie through deregulation, it became rich and powerful enough to fiercely resist being bottled back up.

While regulation and supervision are valuable and necessary, their stringency will loosen and tighten depending on how far we are from the last bad crisis and who controls Congress and the presidency. An FTT, however, will keep the sector moderately sized and in check so long as the tax is not rolled back entirely. By crowding out most financial transactions, it will lead to less trading, and the trading that occurs will cost less. The sector will shrink.

In the absence of effective regulation, a smaller financial sector still might do dumb things that cause crises—but the fallout will be more contained, and public frustration with the sector will be considerably blunted.

Revenue Is Nice, But Base Erosion Is Better

In short, an FTT would do two things: It would raise revenue progressively, and it would shrink the size of the financial sector. These two things will trade off against each other—the more revenue it raises, the less effective it will be at constraining finance, and vice versa. But this is fine. If financial activities continue apace even with an FTT in place, at least the large amounts of revenue raised can support prosocial public spending.

And if revenue comes in beneath projections, this just means more of the financial sector’s activity has been exposed as low-value and then abandoned. This in turn will mean the economy has more resources to put to better use. FTTs should be part of any progressive tax plan. They should also be a high priority for anybody who claims to be seriously concerned about the ill effects of financialization in the U.S. economy.