I – II – III – IV – V – VI – VII – VIII – IX – X – XI – XII – XIII – XIV – XV – XVI – XVII – XVIII – XIX – XX – XXI – XXII – XXIII – XXIV – XXV – XXVI – XXVII – XXVIII – XXIX
[Single-page view]

Now, needless to say, excessive regulation isn’t the only form government overreach can take. In addition to imposing indirect costs on individuals and firms via regulation, government can also impose costs more directly, in the form of taxation. Obviously, as with regulation, you do need to have some kind of taxation if you want to have a government at all. But you can’t just assume that any kind of tax will be fine, any more than you can assume that any kind of regulation must automatically be fine. It matters a lot which particular kind of taxation you decide to implement – because while some taxes can be a net positive, others can most definitely be a net negative (and in fact, even the more positive ones almost always still have some real downsides).

So for instance, let’s imagine that some local factories are emitting harmful pollutants into the air. If you decide to tax those factories for the pollution they emit, in an amount roughly equivalent to the harm their pollution is causing, this will tend to be a good thing overall, because it will actually capture a cost that the normal functioning of the market wouldn’t, and will incentivize the owners of the factories to pollute less. Without the tax, the firms would be free to pollute to their hearts’ content, and the costs of that pollution would simply be borne by the unconsenting public. So implementing the tax is a net positive. On the other hand, if you instead decide to tax other aspects of those firms’ operations, like the products they sell (with a sales tax) or the employees they hire (with a payroll tax), then the net effect becomes more ambiguous – because although you might bring in more revenue, you’ll no longer be disincentivizing the firms from engaging in harmful activities like polluting; you’ll instead be making it harder for them to engage in more beneficial activities like hiring workers and selling products that customers want. And this isn’t just bad news for the firms; it’s bad news for the workers and customers as well.

Milton and Rose Friedman give one example of how this kind of taxation can obstruct the process of hiring new workers:

Employers complain that the wedge introduced by [payroll] taxes between the cost to the employer of adding a worker to his payroll and the net gain to the worker of taking a job creates unemployment.

In other words, if an employer is willing to pay $50,000 a year for a new worker, but 15% of that amount ends up going to the government instead of the worker themselves, then a worker who would have happily accepted any job that allowed them to take home $45,000 a year will instead remain unemployed, because after taxes, their actual take-home pay would only be $42,500. They’re worse off because they don’t get the job, the employer is worse off because they don’t get an employee whose services they were willing to pay for, and most ironically of all, the government doesn’t even get any revenue out of the situation, because without any worker getting hired, there’s no salary from which to draw the tax in the first place. It’s just an outright loss for everyone.

Likewise, the same thing can happen to customers when they have to pay sales taxes on the products they buy. Steven E. Landsburg explains:

Consider, for example, my sandals. I found them on the Internet for $40, though I’d have happily paid $50. There’s a very real sense in which buying these sandals made me $10 richer. Better yet, my gain came at nobody’s expense, so it made the world as a whole $10 richer. That $10 gain—the difference between what I was willing to pay and what I actually paid—is what economists call consumer surplus.

Now, if a sales tax had added, say, $6 to the price of those sandals, I’d still have bought them. I’d be $6 poorer than I am today, but someone else would be $6 richer. So far so good. But if a larger sales tax had added, say, $12 to the price of those sandals, I’d have avoided the tax by not buying the sandals. I would lose my $10 consumer surplus and nobody would win. That’s unambiguously bad.

Even a small sales tax will probably discourage at least a few people from buying the sandals. Their lost consumer surplus is what economists call a deadweight loss because it comes with no offsetting benefit to anyone.

Now, in response to this, you might be tempted to argue that instead of making the customer pay the tax, perhaps we could just impose the tax on the seller after the fact, once the sale has already been made. But in many cases, this approach will simply result in the seller cutting their output and/or raising their prices and passing the cost of the tax on to the customer regardless. In either case, the result will be the same: A lot of customers will miss out on being able to buy the product, either because the price got too high, or because the quantity of the product being sold was reduced and there wasn’t as much of it available to buy anymore – and so these customers will have to settle for a lesser-preferred alternative, and will be worse off. Here’s Joseph Heath on the subject:

The left constantly tries to [claim] that tax burdens can be imposed on corporations rather than individuals. […] Many traditional left-wing parties, for instance, oppose carbon taxes on the grounds that oil companies and polluters should be the ones to pay the penalties, not consumers. But how could that possibly work? When the price of oil goes up, oil companies don’t become less profitable—they simply charge people more for gas. If the taxes they pay go up, their response will be exactly the same. It makes absolutely no difference whether the tax is imposed upon consumers or upon [the] corporations: It will be the consumer who pays.

[…]

None of this should be taken to suggest that there should be no taxes on corporations. It just means that the question is complicated. […] What matters is simply that corporations not be treated as a magic hat out of which the government can pull arbitrarily large tax revenues. Taxes on corporations increase the transaction costs associated with organizing economic transactions outside the market (such as building things in-house rather than outsourcing). This is, in general, undesirable, which is why the most progressive, redistributive welfare states in Europe typically have some of the lowest rates of corporate taxation.

Of course, it can be pretty galling to see corporations raking in billions of dollars and making their owners filthy rich while poorer people struggle, so there’s a natural impulse to want to tax them for all they’re worth. But in light of these factors making it more complicated to tax the corporations directly, one alternative approach might be to leave them alone and instead just raise taxes on their rich owners, after those owners have taken the money out of their companies and there’s no longer the same danger of distorting the companies’ incentives by taxing it. Such an approach might involve imposing an individual income tax, or it might involve some other option like a property tax, a luxury tax, a land value tax, etc. – or it might involve any combination of the above. (The land value tax is my favorite, personally, but that’s a subject for a whole other post.) But regardless of which kind of tax you prefer, the important thing is just to be mindful of the fact that there will always be some cost associated with implementing any tax – if nothing else, simply the basic logistical cost of having to collect and process it. Heath continues:

As economists never tire of pointing out, collecting taxes almost always imposes inefficiencies, by distorting incentives and increasing transaction costs. One of the arguments sometimes made for free public transit is that a significant percentage of each fare is absorbed by the cost of collecting the fare—minting tokens, counting coins, paying attendants, and so on. Standard consumption or income taxes should be thought of in the same way. A nontrivial percentage of each dollar collected in taxes is used to pay the costs of collecting that tax. It is important to be aware of the indirect costs as well, such as the transactions that do not occur because they would be taxed. When people try to put an exact number on this, it tends to be very much influenced by their ideological proclivities, but let’s say, for the sake of argument, that on average it “costs” the government 15¢ to collect a dollar’s worth of tax revenue. This is called the “deadweight loss” of taxation. Because of the deadweight loss, the total cost of any public project will be larger than the dollars-and-cents cost. When the government spends $1, the “social cost” of this is actually more like $1.15. Thus a public project would have to generate at least $1.15 of benefits for every dollar that the government spends in order to be worthwhile undertaking.

In certain cases, this standard can quite easily be met. With public or “club” goods, for instance, where there is an underlying market failure, the state is often the only institution able to provide a particular good. When the government builds roads, provides vaccinations, or operates a judicial system, there is very little doubt that the benefits outweigh the total cost. This is because the state is not just handing money over to people; it is providing them with a good that they would otherwise be unable to obtain (or unable to obtain at a price that would make it worth buying).

With pure redistribution, on the other hand, the government is not providing a good; it is just handing over money. If the government imposed a tax upon Bill in order to give the money to Ted, it would only be in a position to give Ted 85¢ for every $1 in losses imposed upon Bill (speaking roughly). From the perspective of financial cost, it looks like a losing proposition. The only way it could be justified is if Bill were richer than Ted, so that the welfare cost to Bill of losing $1 was less than the welfare gain to Ted of receiving a mere 85¢. This is called a “progressive” redistribution. The important point is that a redistribution, in order to be worthwhile, must be not only progressive, but significantly so, in order to outweigh the losses imposed by the tax system itself. (Using the tax system to do pure redistribution is sort of like using a leaky bucket to bring water from one person to another: You lose a certain amount in transit. As a result, the transfer will be worthwhile only if the recipient is really thirsty compared to the donor; otherwise it’s a waste of water.)

Heath is right to point out that despite the costs, it can still be worthwhile to impose taxes in order to create socially beneficial outcomes. In fact, it very often is. The point here is just to acknowledge that introducing a new tax is always a tradeoff; it isn’t just automatically an unambiguous positive every time. To sum things up, then, here’s a concluding comment from Wheelan:

Government has the capacity to do many good things. [But] even then, when government is doing the things that it is theoretically supposed to do, government spending must be financed by levying taxes, and taxes exert a cost on the economy. This “fiscal drag,” as Burton Malkiel has called it, stems from two things. First, taxes take money out of our pockets, which necessarily diminishes our purchasing power and therefore our utility. True, the government can create jobs by spending billions of dollars on jet fighters, but we are paying for those jets with money from our paychecks, which means that we buy fewer televisions, we give less to charity, we take fewer vacations. Thus, government is not necessarily creating jobs; it may be simply moving them around, or, on net, destroying them. This effect of taxation is less obvious than the new defense plant at which happy workers churn out shiny airplanes.

[…]

Second, and more subtly, taxation causes individuals to change their behavior in ways that make the economy worse off without necessarily providing any revenue for the government. Think about the income tax, which can be as high as 50 cents for every dollar earned by the time all the relevant state and federal taxes are tallied up. Some individuals who would prefer to work if they were taking home every dollar they earn may decide to leave the labor force when the marginal tax rate is 50 percent. Everybody loses in this situation. Someone whose preference is to work quits his or her job (or does not start working in the first place), yet the government raises no revenue.

As we noted [earlier], economists refer to this kind of inefficiency associated with taxation as “deadweight loss.” It makes you worse off without making anyone else better off. Imagine that a burglar breaks into your home and steals assorted personal possessions; in his haste, he makes off with wads of cash but also a treasured family photo album. There is no deadweight loss associated with the cash he has stolen; every dollar purloined from you makes him better off by a dollar. (Perversely, it is simply a transfer of wealth in the eyes of our amoral economists.) On the other hand, the stolen photo album is pure deadweight loss. It means nothing to the thief, who tosses it in a Dumpster when he realizes what he has taken. Yet it is a tremendous loss to you. Any kind of taxation that discourages productive behavior causes some deadweight loss.

Taxes can discourage investment, too. An entrepreneur who is considering making a risky investment may do so when the expected return is $100 million but not when the expected return, diminished by taxation, is only $60 million. An individual may pursue a graduate degree that will raise her income by 10 percent. But that same investment, which is costly in terms of tuition and time, may not be worthwhile if her after-tax income—what she actually sees after all those deductions on the paycheck—only goes up 5 percent. (On the day my younger brother got his first paycheck, he came home, opened the envelope, and then yelled, “Who the hell is FICA?”) Or consider a family that has a spare $1,000 and is deciding between buying a big-screen television and squirreling the money away in an investment fund. These two options have profoundly differently impacts on the economy in the long run. Choosing the investment makes capital available to firms that build plants, conduct research, train workers. These investments are the macro equivalents of a college education; they make us more productive in the long run and therefore richer. Buying the television, on the other hand, is current consumption. It makes us happy today but does nothing to make us richer tomorrow.

Yes, money spent on a television keeps workers employed at the television factory. But if the same money were invested, it would create jobs somewhere else, say for scientists in a laboratory or workers on a construction site, while also making us richer in the long run. Think about the college example. Sending students to college creates jobs for professors. Using the same money to buy fancy sports cars for high school graduates would create jobs for auto workers. The crucial difference between these scenarios is that a college education makes a young person more productive for the rest of his or her life; a sports car does not. Thus, college tuition is an investment; buying a sports car is consumption (though buying a car for work or business might be considered an investment).

So back to our family with a spare $1,000. What will they choose to do with it? Their decision will depend on the after-tax return the family can expect to earn by investing the money rather than spending it. The higher the tax, such as a capital gains tax, the lower the return on the investment—and therefore the more attractive the television becomes.

Continued on next page →