Who says the study of taxation can’t be fun? Most of you, I would expect. But Michael Keen, deputy director of the Fiscal Affairs Department of the IMF, and Joel Slemrod, the Paul W. McCracken Collegiate Professor of Business Economics and Public Policy at Michigan’s Stephen M. Ross School of Business, are out to prove you wrong. Their delightful (you read that right) new book, Rebellions, Rascals and Revenue: Tax Follies and Wisdom Through the Ages, might have been called “Tax Policy Without Tears.” But I like their title better. I won’t try to describe the book further, other than to say it is a sprawling compendium of (mostly weird) anecdotes that neatly illustrate the principles of the economics of tax policy. Here, we excerpt a portion of the chapter titled “Breaking Bad and Making Good,” which is a neatly self-contained essay on how taxes have been used to change behavior. Read it and giggle.
— Peter Passell
Published July 26, 2021
*Copyright Other Press, 2020. All rights reserved
In 1698, Peter the Great launched a series of reforms aimed at modernizing Russia along the European lines he had seen during the boisterous tours of his youth. The construction of St. Petersburg was the grandest of these. Less sublime, but also part of this wider campaign, was his introduction of an annual tax on beards.
The beard tax shows that sometimes taxes are levied not so much to raise revenue as to change behavior, either to discourage the bad or encourage the good. In the jargon, it is referred to as a “corrective” role, which makes it sound less value-laden than what it really is: a form of social engineering.
Do the Right Thing
The use of taxes to induce people to behave as their rulers think they ought to dates back at least to the sumptuary taxes of the ancient world: taxes on luxuries motivated not so much by a desire to tilt the tax burden toward the better-off as by a desire to keep the lower orders in their place. Not so different was the 1928 decision of the Swiss canton of Uri in Switzerland to tax women’s bobbed hair, then the height of fashion. What is regarded as behavior to be discouraged or encouraged varies over time, place and even people, but the instinct to use the tax system to get people to do the right thing remains the same.
Social engineering through the tax system has been rife in relation to marriage and, its close accompaniment, childbirth. Earlier in the book we saw how hard it is to secure horizontal equity [a level playing field] between married couples and single individuals. But that is a very modern concern, and through much of history, marriage has been seen as something to be positively encouraged, largely as an indirect way of encouraging childbirth.
Taxes on bachelors have been one way to do this. In ancient Greece and Rome, the aes uxorium was levied on unmarried men after the normal age of marriage. The Ottomans also had an annual tax on bachelors, and so too, did the British between 1695 and 1706. In the United States, bachelor taxes surfaced in colonial times and continued into the 20th century: Georgia, Maryland, Montana and Texas all at one time had bachelor taxes.
Proponents of bachelor taxes offered many justifications. “Rich bachelors should be heavily taxed,” declared Oscar Wilde: “It is not fair that some men should be happier than others.” A more standard argument (though to much the same effect) was that bachelors of a given income and wealth had more resources than did married men, as they had no families to support, so had a greater ability to pay.
Encouraging childbirth is a recurrent objective. It has often reflected a militaristic desire to build up numbers relative to rivals: Fascist Italy and Nazi Germany both had bachelor taxes (and other generous support for marriage and childbearing). Protecting the position of favored races by expanding their numbers has also sometimes been a motive: In South Africa, the Transvaal had a levy of this kind from 1917 to 1920, allegedly to encourage white procreation to keep up with a burgeoning black population.
Tender-hearted policymakers did sometimes worry about those who were unlucky in love. Were they to be doubly cursed, embraced by the taxman but spurned by womankind? So they sometimes exempted bachelors who could prove that they had asked a woman to marry them, only to be rejected. It was a wise man who made his proposals with a ring in one pocket and a tax waiver form in the other.
In Argentina circa 1900, such a provision gave rise to one of history’s more bizarre forms of tax avoidance. For a small fee, “professional lady rejecters” agreed to swear to the authorities that a particular man had indeed proposed to them and that they had declined the offer.
Taxing bachelors encourages marriage, but not necessarily childbirth. If childbirth is the objective, then it is best to reward it directly, the indirect route having the presumably unintended consequence of producing marriages that are unhappy, childless or both.
This line of reasoning is just one instance of a general principle of tax design: if the idea is to encourage (or discourage) something, it is best to use the tax (or subsidy) most directly focused on that end.
This application of the principle of targeting was implemented from the 1940s in Stalin’s Soviet Union and other communist countries, with the imposition of a tax on childlessness designed to boost the birth rate and recover from wartime population losses.
In the Soviet Union, an additional 6 percent income tax applied to childless men from age 25 to 50 and to childless married women from 20 to 45. The levy in Romania, called a “celibacy tax” — with the happy connotation that having sex was being encouraged — was imposed on anyone still childless by the age of 25.
The same principle was applied in Mao’s China, but the other way around. There, from the late 1970s and early 1980s (and until 2016), tax-like measures were used to actively discourage families from having more than one child. Although well-targeted as a fiscal measure, unintended consequences kicked in — a regular occurrence with almost all taxes. Given a traditional preference for male offspring, there emerged a massive imbalance between the sexes: by 2014, there were 32 million more boys than girls in China.
The use of fiscal incentives to encourage childbirth may sound archaic, a remnant of times when a country’s military power depended in large part on its sheer population. But it is making a comeback, as many countries struggle with the aging of their populations and the consequent difficulty of financing state-provided pensions and health care for the elderly.
Although well-targeted, unintended consequences kicked in — given a traditional preference for male offspring, there emerged a massive imbalance between the sexes.
Going beyond the usual tax breaks or cash payments for child support, Australia, Canada, the Czech Republic, Lithuania and Singapore all provide a baby bonus. In 2019, Hungary introduced a lifetime income tax exemption for women with four or more children. And some of these policies do seem to affect births — or at least their timing. In 2004, the Australian government announced that parents of children born on or after July 1 that year would receive a $3,000 one-time dependent exemption. Sure enough, the number of births dipped sharply before July 1, and then, on July 1 itself, more Australian children were born than on any other day in the previous 30 years. Changes in the timing of induction and Cesarean section procedures accounted for most of the extra deliveries.
The childbirth example also shows the similarity between taxing bad things (not having children, in Romania) and subsidizing, or giving a tax break, to the good thing that is its opposite (having children, in Australia). Peter the Great could have had much the same impact by instead handing money to beardless boyars, and giving tax breaks for marriage creates much the same incentives as does taxing bachelors. The key advantage of taxing bad things rather than subsidizing good ones is, of course, that it provides government with additional resources rather than using them up. Revenue concerns, even if they are not the main motive for these policies, can never be entirely ignored.
When enacted in 1712, the British tax on newspapers (and the paper they were printed on) was just one in a long list of revenue-raising excises. As the popularity of the press grew, it came to inspire some impressive avoidance responses. Newspapers could be rented by the hour, were passed from post office to post office and circulated in ale houses and coffee houses. But by the early 19th century, revenue was clearly not the only purpose of the tax. The government’s suppression of some parts of the press was explicit, with the censors’ language in 1819 referring to pamphlets and papers that “excite Hatred and Contempt of the Government.”
The publisher of two local “respectable” newspapers argued that removing the newspaper tax would “effect a revolution in the character of the nation” due to the infiltration of men of lower standing into the publishing business. He also contended that repeal would open the floodgates for cheap journals of dubious quality, depriving the more respectable journals of advertising revenue.
William Gladstone abolished the paper duty in 1861 — over the objections of Minister Lord Palmerston, who thought the idea of suppressing a mass press rather a good one. So did Queen Victoria.
Social reformers saw things differently, arguing that the tax blocked working households’ access to news and information. Showing an eye for a good (and enduring) slogan, they dubbed it the “tax on knowledge.” Chancellor of the Exchequer William Gladstone agreed, and abolished the paper duty in 1861 — over the objections of Prime Minister Lord Palmerston, who thought the idea of suppressing a mass press rather a good one. So did Queen Victoria, and the patrician House of Lords went so far (before being brought in line) as to violate the convention that it could not reject tax measures by voting them down.
One historian identifies the freedom of the press from taxation as the most important driver of changes in newspaper production in the 19th century. But the press was not free from taxation everywhere. The Austrian government of the late 19th century still felt much the same as Lord Palmerston. A correspondent of the New York Nation reckoned in 1897 that the newspaper tax there was responsible for “the extreme ignorance of the lower classes as regards all political, social and industrial questions of the day. Indeed, this is the sole object of the reactionary party in imposing such taxation.”
The United States has usually refrained from taxing newspapers, or the press generally, perhaps as a lasting consequence of the hated Stamp Act of 1765 imposed by Britain on all legal documents and printed materials, including newspapers. But it has been tried. In 1934, Louisiana, under the thumb of Huey Long, passed a tax on advertising sales for newspapers with a circulation exceeding 20,000 (and so applying to the urban newspapers that opposed him). Long called it a “tax on lying,” presumably an archaic term for fake news. In 1936, however, his tax was unanimously rejected by the U.S. Supreme Court as violating freedom of the press.
These days, many countries bend over backward to favor, rather than penalize, the media (now including digital forms). All but two EU members provide preferential VAT treatment to printed books, with the United Kingdom and Ireland going furthest and applying a zero rate. From December 2020, the UK government removed VAT on e-books and online newspapers, which opponents had labeled the “reading tax.” In 1983, the Supreme Court rejected a Minnesota law taxing paper and ink products used by newspapers. As in many other areas of tax policy, Gladstone has prevailed.
Tax Bads, Not Goods
Although most slogans about tax design are untrustworthy, this one is hard to argue with. It is much easier, however, to identify what is bad than to put numbers on just how bad it is and therefore exactly how heavily it should be taxed. But economists have at least come up with a clear way of thinking about these things.
At the heart of their approach is the idea of an “externality”: a damage (or benefit) that a transaction or action confers on those who have no say in whether it takes place — and whose interests are therefore presumably ignored by those who do have a say. Taking for definiteness something that generates a harmful externality, like pollution, this means there will be too much of it, in the sense that the benefit to those generating the externality from the last bit of pollution (which might be, for example, the expense spared by not using a cleaner technology) is less than the cost imposed on those suffering from it.
In principle, this difference leaves scope for mutually beneficial bargaining between polluter and pollutee: both sides would gain if those polluted were to induce the polluter to pollute a little less by paying them an amount larger than the expense of cutting emissions a bit, but less than the harm they would suffer from those emissions. That is a happy outcome, because pollution then ends up at an “efficient” level, in that there is no way to change it without reducing the net benefit enjoyed by the two sides taken together. The externality is fully dealt with.
That is a neat solution, but in practice, externalities commonly affect so many people that this kind of direct bargaining between polluters and pollutees is impossible. Taxes, however, can also do the trick, as first seen a century ago by the Cambridge (UK) economist Arthur Cecil Pigou: all that is needed is to tax the damaging activity at a rate equal to the monetary value of the damage that the last burst of that activity, beyond the efficient level, causes to others. In this way, the damage caused to those others will turn up as a cost to be taken into account, however selfish they are, by the polluters.
Ever since Pigou, the recommendation that taxes be deployed to correct for (“internalize”) externalities has been a standard part of the economist’s tool kit.
So compelling is the case that, ever since Pigou, the recommendation that taxes be deployed to correct for (“internalize”) externalities has been a standard part of the economist’s tool kit (the only disagreement being whether such taxes should be described as Pigouvian or Pigovian).
Polluting activities are the classic example of a negative externality — a laundry, for instance, that dumps dirty water into a river causes damage downstream to other businesses or swimmers. A Pigovian tax would charge that laundry an amount per liter of dumped water equal to the monetary value of the harm those others suffer. (This general idea is sometimes referred to as the “polluter pays principle,” which alas falls into the category of slogans not to be trusted.) But externalities can also be positive. For example, basic scientific progress can generate positive externalities by enabling other researchers to develop further advances. These cases call for a Pigovian subsidy.
Over the years, generations of bored students have had to sit through cheesy examples of externalities, like our laundry one, to illustrate the idea of Pigovian taxation. Now, however, we have a very big and bad real-life externality to which to apply the idea: climate change.
Saving the Planet
Scientists pretty much agree that the accumulation of greenhouse gases — about 65 percent of which consist of carbon dioxide (CO2) released by burning fossil fuels (oil, gas, coal) — is increasing average global temperatures by trapping heat reradiated from the earth’s surface. The consequent change in climate patterns implies substantial and generally bad economic effects: more, and more damaging, extreme weather events, much lower output in many low-income countries, and increased risk of Day After Tomorrow-like catastrophic events, such as a reversal of the Gulf Stream or collapse of the West Antarctic ice sheet.
The Pigovian response to this “mother of all externalities” is straightforward: tax greenhouse gas emissions in general, and emissions from burning fossil fuels in particular, at a level that reflects the global damage they will cause. Because the amount of CO2 released in burning any fossil fuel is proportional to its carbon content, such a carbon tax is, in principle, easy to implement. Just figure out how much damage an additional CO2 emission causes and charge each unit of fossil fuel that amount for each unit of CO2 that its burning produces. (And do so similarly for other greenhouse gases.) Some or all of this tax will be passed on to the consumers of goods whose production uses carbon-based fuels, reducing demand for those goods. The rest will be passed back to those selling fossil fuels, leading them to reduce supply.
In this way, individuals and businesses will be induced to cut emissions to reflect the harm they do — not necessarily to stop emitting entirely, but to balance that harm against the costs of reducing emissions. And research and development into, and investment in, low-carbon energy technologies that avoid the tax would be spurred.
Technically, the hard part is gauging the level at which the carbon tax should be set. In the Pigovian logic, it should be the social damage caused by the last CO2 emission. Views differ on exactly how large that marginal damage is, and even advocates of carbon taxation acknowledge “there’s no magic formula or perfect number.” But a reasonable ballpark figure, as of today, might be a charge of $35 per (metric) ton of CO2, which comes to about 31 cents per gallon of oil.
This may not sound like much, and, indeed, it would be swamped by the swings in gasoline prices that we have all become used to. However, the biggest issue is not gasoline, but coal, which is dirty (that is, CO2-intensive in its production of energy), widely used and available in vast quantities.
A carbon tax takes us back to the future: The first coal tax in England was introduced in 1368, and a charge on coal entering London continued until 1889.
In this respect, a carbon tax takes us back to the future: the first coal tax in England was introduced in 1368, and a charge on coal entering London continued until 1889 (funding, among other things, the rebuilding of St Paul’s Cathedral after the Great Fire of 1666). For coal, carbon taxation is a very big deal. That same carbon tax of $35 per ton of CO2 could roughly double coal prices. And, to have the intended effect, the charge on burning carbon should rise steadily over time, faster than prices in general, with a tax of something like $75 per ton needed by 2030 if warming is to be contained within the limits aspired to in the 2015 Paris Agreement.
Importantly, there is another way of achieving the same effect as a carbon tax, illustrating the occasionally thin line between tax and other policy instruments. To see this, suppose that with a tax of $50 per ton of CO2 emissions would be 30 billion tons. Viewed the other way around, that means that the right to emit a total of 30 billion tons could be sold for a price of $50 per ton. Hence the alternative: create rights to emit 30 billion tons of CO2, sell them and allow private markets to trade in them. The price would then settle at that same $50 per ton.
Such an emissions trading system — also known as “cap and trade” (because there’s an upper limit to emissions, and because rights can be bought and sold) — can in principle replicate the effects of a carbon tax and raise the same revenue for the government if the rights to emit are sold at auction.
Most of the world has recognized the need to reduce carbon emissions, with nearly 190 nations plus the EU having ratified the landmark Paris Agreement of 2015, almost all of them submitting quantified pledges for their own mitigation strategies. And there is remarkable consensus among economists, of all political leanings, on the importance of carbon pricing as the most efficient and effective way in which to meet these commitments (and, before too long, go further).
But the world remains far from fully embracing the idea. Although there are about 60 carbon taxes or emissions-trading schemes in place — including, for example, a muchadmired carbon tax in British Columbia and an EU-wide trading scheme — the average price on global emissions is only about $2 per ton. For the inconvenient truth is that establishing meaningful carbon pricing often encounters powerful resistance. Governments are reluctant to act unilaterally for fear of making their firms less competitive in world markets.
Countries and companies sitting on large deposits of fossil fuels worry that their assets will be stranded, losing value as emissions are taxed. Consumers do not like higher energy prices. Low-income countries wonder why they should make it harder for themselves to ensure decent access to energy for their poor in order to solve a problem created by the rich countries.
There are ways to address these issues — most of them, at least.
Transfer payments can be used to protect the poorest from higher fuel prices and fossil- fuel-producing communities subject to displacement. And there are ways to price the use of carbon as an input so as to have similar effects on emissions without having such a large impact on the prices of what it is used to produce. Armed with the economics of Pigovian taxation, climate change is not the biggest intellectual puzzle mankind has been presented with. The problem is an unwillingness to embrace this key part of the answer.