And the same story could be told about what governments spend money on. In 1910, they provided public goods such as national defense, law and order, street lighting, and public hygiene. They still do. But they now also spend vast sums on supplying private goods such as education, health care, pensions, and unemployment insurance, to name just a few big ones. In 1910, U.S. government spending on health care was 0.3% of GDP. By 2019, it was 8.1%, greater than total government spending in 1910. U.S. government spending on pensions has risen from 0% to 6.8% of GDP; on education, from 1% to 6%; and on welfare, from 0.2% to about 3% (on average through the economic cycle).
Politicians on both the political left and right have brought about this growth. Yet, some politicians—sometimes the very ones increasing government spending—have lamented it. An unfashionable few still do. These dissenters have typically argued that it can’t be afforded. More spending will push taxes and government debt too high. Soon enough, the government will find itself in the position of a household that has maxed out its credit cards and has no way of increasing its income.
Enthusiasts for government spending accuse these worrywarts of committing the “household fallacy.” Governments differ from households in ways that mean they do not face a budget constraint—at least, not one that the U.S. government has ever been near to bumping up against. Most importantly, governments can print money to pay their bills. Householders who try to do that go to prison. Modern Monetary Theory is the culmination of this line of thinking.
From the absence of an effective budget constraint, many move directly to the view that government spending should be increased. If you can afford a nice house and a Tesla and a yacht and foreign vacations, why not buy them? Similarly, if the government can afford to improve citizens’ lives by supplying more or better health care, education, income insurance, roads, museums, television channels, and all the rest of it, why not?
Those who think this way are committing their own household fallacy. It may be true that a household should buy whatever it both wants and can afford. But this isn’t true of governments. Even if government spending can be afforded, it’s a bad idea because of an important way in which nations differ from households. Politicians are not to citizens what parents are to their children. Whereas parents know and love their children, politicians neither know nor love the citizens on whose behalf they direct government spending. And whereas parents voluntarily spend their own money on their children, politicians spend money extracted involuntarily from citizens. These facts mean that government spending on private goods is simply an expensive way of forcing people to consume things.
Let’s Play “Spend for Me”
Imagine you were a contestant on a game show called Spend for Me. The goal is to allocate the spending of people you do not know as they themselves would choose to—say, 30% on housing, 15% on food, 5% on clothes, and so on.
The task would be relatively easy if you were spending on behalf of a pauper. He would put all his income into the basics: food, shelter, and clothing. But imagine you were spending for a billionaire. You wouldn’t know where to start. Or, rather, you would know only where to start. The basics consume less than 1% of his income. How would you allocate the other 99%? Should you buy a football team? A Picasso? Caseloads of Château Palmer 1961? Everyone buys roughly the same things with their first $10,000 of annual income, but then they start to show some individuality.
Politicians put themselves in the position of contestants on Spend for Me. Insofar as they mean to promote the welfare of the population, they must allocate government spending on behalf of other people in the same way that those other people would allocate it for themselves. (Some politicians deny this, of course, but we’ll get to that later.) In a country where annual GDP per person is less than $1,000, such as the Republic of Congo, this task might be relatively easy. But when annual GDP is $70,000 per person, as in America, the task is impossible. Yet, perversely, the U.S. government did not attempt it in 1910, when GDP was $10,000 per person (in today’s money), but does it now, when per-capita GDP is seven times higher.
The inevitable misdirection of consumption is a bigger problem for the poor than the rich. Suppose the government spends $20,000 on your behalf—allocating, let’s say, $8,000 to a school for your child, $2,000 to unemployment insurance, $4,000 to health insurance (subsidies), $4,000 to retirement savings, and $2,000 to subsidies for opera, food safety certification, and all the other little bits and pieces the government provides. If the politicians have got your preferences wrong and you consume $50,000 a year, that’s a serious problem. The government is misdirecting a large portion of your total consumption.
In contrast, if you consume $500,000 a year, misdirection within the government’s $20,000 is a negligible problem. For simplicity, suppose politicians allocate $10,000 to things you value at only $5,000—that is to say, the government spending makes you $5,000 worse off than you could have been. Given the diminishing marginal utility of consumption, this $5,000 loss is far greater if you consume $50,000 than if you consume $500,000. Indeed, on the standard assumption that the marginal utility of a dollar of consumption is inversely proportional to total consumption (income), the loss is 10 times as great.
The misallocation of spending occurs not merely between categories—health insurance vs. education vs. retirement saving vs. whatever else—but within them. The government supplies a school to which you may send your child without paying fees. Even if, by some miracle, the per-student cost of the school matches what you would have chosen to spend, why should it be a school of the sort that you would choose to send your child to?
Again, the problem is bigger for the poor than the rich. People who earn $500,000 a year can afford to throw away the education they have already paid for through their taxes and buy an alternative they prefer. Someone who earns $50,000 cannot. In the UK, the government supplies medical treatments deemed to be worth the cost not by consumers but by a government agency, the National Institute of Clinical Excellence (NICE). The rich can afford to buy health insurance policies that better suit their preferences, having already paid for the government policy through their taxes. Most Brits cannot. They must take the health insurance policy the government provides, one that includes payment by queuing. The individuals who get the treatments deemed worthwhile by NICE are those who can get to the front of the queue before they die.
The whole business is quite peculiar. Little could be easier than getting each individual’s consumption to match her preferences: let her decide for herself. Taxing the population and then using the money to give people private goods is a recipe for failure. What could possibly justify it?
Entitlements
A popular justification for the governmental supply of private goods is that citizens are somehow entitled to them. The government must supply what people have a right to!
To see the error in this justification, imagine someone knocked at your door and presented you with a laptop computer. You ask how much he wants for it. “It’s free,” the visitor explains. “Everyone is entitled to a laptop. Social justice be done!” As he turns to leave, he adds, “By the way, I bought it with money I got by hacking your bank account.” Justice be damned! You were going to use that money for a new suit. Giving you what you are supposedly entitled to has harmed you.
Or imagine the same situation, except this time he adds, “By the way, I bought it with money I got by hacking your neighbor’s bank account.” Even if your neighbor owes you money and you have no qualms about his loss, you might still regret receiving the free laptop instead of the cash. After all, you would have preferred a new suit, which you could have bought with the cash.
Perhaps you really are entitled to a laptop. But this entitlement must surely be interpreted as meaning that anyone who wants a laptop should have one. Politicians could provide this entitlement by ensuring you have enough money to buy a laptop if you want one. If, instead, they provide you with a laptop, they effectively compel you to buy one. It is not an entitlement but a tax-funded obligation.
The beneficiaries of this state provision are not those who receive “free” laptops. If they had wanted one, they could have bought it with the money confiscated from their bank account or from their neighbor’s. The beneficiaries are the suppliers of the laptops that our hacker compels people to buy. They make “sales” they would not otherwise have made. The victims are the recipients of the “gift” who miss out on buying things they would have preferred to the laptop, as well as the producers of those other goods and services, who miss out on sales they would otherwise have made.
It is no surprise that doctors organizations incessantly lobby for increased government spending on health care and that school teachers and university professors lobby for more government spending on education. Everyone has a right to excellent health care and education! Or, in other words, people should be forced to spend more on what doctors and teachers supply. I have even heard opera singers claim that everyone has a right to affordable tickets to the opera and, therefore, that the opera companies that employ them should be subsidized from taxation. And, indeed, opera is subsidized in many countries. The fashion for entitlements has really come to something when it can be used to force people who prefer death metal to pay for opera.
Politicians as Parents
Like anything else, consumption is good for you if its benefits exceed its costs. What’s a benefit and what’s a cost, and the size of that benefit or cost, depend on your preferences. Is the increased chance of waking up in bed with a stranger a benefit or a cost of going out and drinking heavily? That depends on how you feel about the eventuality. For some, it’s a cost of drinking heavily; for others, it’s a benefit.
Preferences are an imperfect guide to personal welfare because we humans are fallible. We don’t know all the effects of the things we might do or the outcomes we seek. We can go wrong by overestimating the benefits of something or underestimating the costs. That’s why parents often make decisions on behalf of their children. The little darlings are too ignorant about the likely effects of their actions. A caring mother doesn’t let her child eat the whole bag of candy at once. She knows that her child doesn’t understand the sickening effect.
This is how some see the relationship between politicians and adult citizens. As behavioral economists are keen to remind us, we adult humans are prone to predictable errors. Politicians can help us by making us do what we would do if only we were not error prone.
Though fashionable, this is a preposterous idea. When it comes to knowing what we should do, each of us has a massive informational advantage over politicians. Unlike the politician, you know your particular preferences and your particular circumstances. It is impossible to know what someone should do when you know neither of those things. A parent can make good decisions on behalf of her child because she does know these things. A politician cannot make a good decision for distant adults because she does not know them. Even if politicians were intellectually superior to other people, and even if they wanted what is best for each of us, this unavoidable ignorance means that transferring our consumption decisions to them is a mistake. And, of course, they aren’t in fact cleverer than the rest of us and they don’t genuinely love us. Politicians are not to adult citizens what parents are to their children.
Deadweight Loss
Another difference between parents spending for their children and politicians spending for citizens is that the former spend their own money but the latter do not. Government spending is funded by taxing the population on whose behalf the spending occurs. It therefore entails the deadweight loss caused by taxation.
Start with corporate taxation. Suppose that the cost of equity capital for Acme Kitchen Gadgets is 15%—that is, given the risk it presents, Acme can attract investors only if it offers a return of at least 15% on equity. Acme seeks capital to launch a new device that their best estimates suggest will deliver a return of 18%. That makes it a worthwhile venture. They should proceed. But the 21% corporate tax means investors will get a return of only 14.2%, which is below their 15% “hurdle rate” for investment. Acme won’t get the capital and the worthwhile venture won’t go ahead. The corporate tax prevents worthwhile ventures from being undertaken by driving a wedge between what the venture yields and what the investors receive. The value of the forgone output is the deadweight loss. The bigger the tax, the bigger the wedge, the greater the forgone output, and the greater the loss to society.
The same occurs with other types of taxes. Suppose Jill is willing to pay up to $15 an hour for someone to work in her garden, and Jack is willing to do the work for anything above $14 an hour. They should be able to come to a deal from which both would benefit. But a 10% income tax would mean Jack would receive no more than $13.50 even though Jill is willing to pay $15, and Jack won’t do the work for that wage. Similarly, suppose Jack is willing to pay up to $15 for a widget that Jill is willing to sell for no less than $14. Again, they should be able to do mutually beneficial business. But if there is a 10% sales tax, Jill would receive no more than $13.50 for the widget even though Jack is willing to pay $15 for it, and Jill won’t sell for that.
How great is the deadweight loss from taxation? What is the value of the opportunities for valuable activity that are forgone? The answer depends on the type of tax and on its level. And, even taking these matters into account, economists find the loss difficult to estimate. Given the current mix and level of taxes in the United States, estimates for the deadweight loss from taking an additional dollar from people who are already top-bracket income taxpayers (which is where most politicians now seek extra tax revenue) range from material but smallish to very large.
We don’t need to know the exact loss to see the folly of taxing people to provide them with private goods. Suppose the deadweight loss is 20¢ for an extra dollar of tax, which is at the low end of the range of estimates. A private school that charges fees of $15,000 a year costs society $15,000 per student. A public school whose tax-funded budget is $15,000 per student costs society $18,000 once the deadweight loss is included. Taxing people and then providing them private goods, such as education, is not just a pointless money-go-round that prevents people from consuming as they would prefer; it also increases the cost of supplying those unwanted goods.
Conclusion
Many people believe that the government should guarantee citizens some minimally decent standard of living. Maybe so. But this could be achieved by transfers of cash to those whose lives would otherwise fall below this standard. In Western societies as they are today, this could be achieved by transfers equal to about 10% of GDP. With government spending on public goods (such as law and order and national defense) requiring about 5% of GDP, total government spending and taxation would be about 15% of GDP. The demands of “social justice” would be satisfied, and the gains in consumer welfare and economic output would be huge.
Why, then, don’t politicians offer this simple transfer program and voters plump for it? That is a question for experts in public choice. In some cases, such as opera subsidies and state-funded university courses in post-structuralist hermeneutics, it is probably explained by the lobbying of concentrated groups of suppliers who enjoy the benefits, along with the apathy of the far greater number of taxpayers across whom the costs are dispersed. More generally, promising to spend more on this or that alleged entitlement may be a good way for politicians to disguise what would otherwise be blatant vote buying. “Everyone deserves the best possible health care!” sounds better than “Vote for me and I will give you an extra $5,000!” And there are surely ideological interests at work. Despite its implausibility, many people are convinced that they know how other people should live and, therefore, what they should consume. The kinds of people who enter politics are especially prone to this self-aggrandizing delusion.
Whatever explains the massive growth of government spending over the last century, the effects are dreadful. There’s less fruit in the bowl, and people have less control over whether they get an apple or an orange or a rambutan. Those who call for increased government spending invariably present themselves as caring more for the population than those who resist it. Maybe they do. But, if so, they merely confirm the Mills Brothers’ thesis that you always hurt the one you love.
