Republican tax cuts side by side, from Harding all the way to Trump

There are two kinds of people who talk about tax cuts, and I want to introduce you to both, because you have met them at every holiday dinner. The first says that every tax cut in history has been a gusher of new revenue, that the government actually collects more money when it asks for less, and that anyone who disagrees has never run a business. The second says that Republican tax cuts have only ever been gifts to the rich, that they blow up the deficit, and that anyone who disagrees has never read a budget.

Folks, both of them are working from slogans. And the slogans fall apart the moment you put the actual laws on a table and look at the actual rates. That is what I propose to do here. Eight laws, a century, and a few rows of numbers. I will give you the conservative case at its best, then the objection at its best, and then I will tell you where I come down.

The table

First, a note on method. Tax rates have two parts that people confuse. The top marginal rate is the rate on the last dollars earned by someone at the highest bracket. It is not what anyone pays on all of his income. The corporate rate is the top rate on a company's profits. Both numbers change what it costs to earn one more dollar, which is why economists watch them. Everything below is from the statutes themselves, and you can check each one at Congress.gov or the Internal Revenue Service.

A word about the 1920s corporate numbers: the rates in that decade were in the low teens and moved slightly from act to act. Treat the figure in the table as approximate, and check the statutes if you need a decimal. The individual rates are the ones the argument turns on.

What a rate means: the arithmetic of one more dollar

Here is the part nobody puts on the bumper sticker. A marginal rate is the share of the next dollar that the government takes, so one minus the rate is the share you keep. Do that subtraction down the table and the argument for the cuts becomes plain.

• At the 73 percent rate of 1921, you kept 27 cents of the next dollar. At 58, you kept 42.
• At the 1925 rate of 25 percent, you kept 75 cents. That is 75 divided by 27, or about 2.8 times what you kept before.
• At 70 percent in 1981, you kept 30 cents. At 50, you kept 50. That is 50 divided by 30, or about 1.7 times as much.
• At 50 in 1986 you kept 50 cents, and at 28 you kept 72. That is 72 divided by 50, or 1.44 times as much.

This is the logic behind every one of these laws. The conservative claim is not "the government collects more." The conservative claim is that people respond to what they keep, that work, investment and risk are decided at the margin, and that a code which confiscates three-quarters of the next dollar pushes people to arrange their affairs to avoid it. It is a claim about behavior, and it is a strong one. Ronald Reagan, a former movie actor, liked to argue that at the highest brackets there was little reason to take on one more job. It is a plain argument, and it was true enough in the old high-rate years that tax avoidance became an industry of its own.

Three different designs

It is a mistake to treat these laws as one thing. They are three families, and they work differently.

The 1920s: pay down the war. Treasury Secretary Andrew Mellon, who served under Harding, Coolidge and Hoover, argued that very high rates on the largest incomes were driving capital into tax-exempt bonds and away from productive investment, and that lower rates would bring the money back into the taxable economy. The 1921, 1924 and 1926 acts cut the top rate from 73 to 25 over five years. At the same time, the government ran a budget surplus in every year of the decade, and the national debt, which stood around 25.5 billion dollars at its wartime peak in 1919, fell to around 16 billion by 1930. (Those figures are the standard Treasury numbers rounded; check them against the Treasury's own historical tables.) The cuts were paired with spending restraint under Harding and Coolidge, a story told in The Mellon Tax Cuts of the 1920s and What Happened to Revenue and Silent Cal said little and ran one of the loudest budget records in history. The key point is the pairing. The tax cuts did not pay the bills. The surpluses came because spending was brought down from wartime highs.

The 1980s: lower rates, fewer loopholes. The 1981 act cut every individual rate and, beginning in 1985, indexed the brackets to inflation so that a raise that only kept up with prices would not push you into a higher bracket. That end of "bracket creep" is arguably the most lasting piece of the law; our piece on reagan's 1981 tax cut and the quiet end of bracket creep explains it. The 1986 act went in a different direction: it cut the top rate to 28 percent and closed a long list of shelters and special deductions, so that lower rates applied to a broader base. It was bipartisan, and it was the kind of reform that comes once in a generation. The details are in the 1986 tax reform that lowered rates by closing loopholes.

The 2000s and 2010s: targeted cuts with expiration dates. The 2001 and 2003 laws cut rates, created a 10 percent bracket, expanded the child credit and lowered the tax on dividends and capital gains to 15 percent. They were written to expire at the end of 2010, because of Senate budget rules that limit how long a reconciliation bill can add to deficits beyond the budget window. (More in The 2001 and 2003 Bush tax cuts and why they came with an expiration date.) The Tax Cuts and Jobs Act of 2017 cut the corporate rate from 35 to 21 percent, a permanent change, roughly doubled the standard deduction, and cut most individual rates, with the individual provisions written to expire after 2025. For where those individual provisions stand, do not rely on a blog; check the Internal Revenue Service and Congress.gov. Our piece on five changes in the 2017 tax law that reached the kitchen table explains the changes that reached ordinary households.

The claim: "tax cuts pay for themselves"

Now to the slogan, and I am going to tell you something that will make some of my friends uncomfortable. The claim as usually stated is wrong. And the best conservative economists never made it.

Here is the strongest version of the claim. Arthur Laffer's famous curve says that at a 100 percent tax rate, the government collects nothing, because nobody will work for nothing, and at a zero percent rate it collects nothing, because it does not tax. Somewhere in between there is a rate that raises the most revenue, and above that rate, cutting the rate raises revenue. That is true as a statement of logic. The question is where the peak is. If the real top rate is at 73 percent, as in 1921, the peak is plausibly below it. If it is at 37 percent, as in 2018, it is very likely above it. Laffer himself, as our explainer The Laffer curve explained, and the claim it never actually made shows, never argued that every cut pays for itself.

Here is the case against the slogan, in numbered points.

1. The official scorekeepers say the growth recovers part of the lost revenue, not all of it. The Congressional Budget Office and the Joint Committee on Taxation both run models that include the effect of a tax cut on growth and behavior, and they generally find that those effects offset some of the cost. (I am describing the direction, not quoting a figure; their reports are public and you should read them.) Conservative economists dispute the size of the effect, and that is a fair argument. But "offsets some" is a very different sentence from "pays for itself."
2. Nominal revenue grows anyway. Prices rise and the economy grows, so federal revenue in dollars goes up almost every year after almost any tax law. A chart showing revenue rising after a cut proves very little. The right comparison is revenue against what it would have been without the cut, and nobody can observe that.
3. The 1920s surpluses came from spending restraint. The pattern above is the honest one: tax cuts plus spending cuts.
4. Tax cuts without spending restraint produce deficits. That is arithmetic. If the government collects less and spends the same or more, it borrows the difference. The party that cuts taxes and refuses to cut spending has done half a job. See Why Spending Restraint Keeps Losing in Washington, Whoever Is in Charge.

So where do I come down? I come down on the side of the cuts, for the reason I gave above: they let working people and employers keep more of the next dollar, and that matters even when it does not pay for itself. But I make that argument as a value claim and a behavioral claim, not as an accounting claim. The honest conservative position is: lower rates are good because the money belongs to the people who earned it and because lower rates grow the economy, and we should pay for them with restraint on the spending side. If we say instead that cuts always pay for themselves, we hand the other side an easy win, because anyone with a spreadsheet can show that is false.

The objections from the left, answered

The strongest objections deserve straight answers.

"The cuts favored the rich." The top-rate cuts did reduce the rate paid by those at the top, and no honest person denies it. But the distribution is more complicated than the headline. The 1986 act removed millions of low-income workers from the income tax rolls through a bigger personal exemption and standard deduction, and the 2001 act created the 10 percent bracket and expanded the child credit. Another fact: after the 1920s cuts, Treasury data showed the top earners paying a larger share of total income tax, although that claim comes from Mellon's own department and historians still argue about it. Check the numbers at the source before using any of this.

"The deficit." True: the 1981 act and the 2001 and 2017 laws all increased deficits in the near term, and the 1981 experience is the reason Reagan signed tax increases in later years. The Tax Equity and Fiscal Responsibility Act of 1982 raised revenue, and in 1990 George H. W. Bush agreed to a deal that raised taxes, a reversal that cost him politically; the story is in Read My Lips and How the 1990 Budget Deal Led to the 1992 Defeat. These were not conservative victories.

"The economy did better under other policies." Other periods, including the tax cut of 1964, proposed by President Kennedy and signed by President Johnson, which cut the top rate from 91 to 70 percent, show that the principle is not partisan. Our piece kennedy cut tax rates too, and what the 1964 law proves makes the point. A conservative should be happy to claim that the idea works when a Democrat does it, too.

A rule of thumb for judging any tax cut

When a politician from either party tells you what a tax law will do, put four questions to the proposal.

1. What happens to the marginal rate, not just the average? That is the rate that changes behavior.
2. Is the base broadened or narrowed? Lower rates on a broader base (1986) are a different animal from lower rates with new loopholes.
3. Is it permanent or does it expire? Temporary cuts change timing; permanent ones change plans.
4. What does it do to spending? A tax cut with no spending discipline is a loan from your grandchildren.

The bottom line

• Eight Republican laws, one table. The top individual rate fell from 73 percent in 1921 to 25 percent by the mid-1920s, from 70 to 50 in 1981, from 50 to 28 in 1986, from 39.6 to 35 percent in the 2000s and from 39.6 to 37 in 2017. The corporate rate fell from 46 to 34 in 1986 and from 35 to 21 in 2017.
• The conservative case is about what you keep of the next dollar: at 73 percent you kept 27 cents; at 25 percent you kept 75. People work and invest at the margin.
• "Tax cuts always pay for themselves" is wrong, and Laffer never said it. The cuts of the 1920s worked alongside spending restraint. Cut taxes and spending together, or expect to borrow.
• Before you argue about any tax law, read the statute's own text or the IRS summary and look at the rates yourself. For the 2017 individual provisions, check the current status at Congress.gov and the IRS.
• Ask your representative one question: if you cut my taxes, what are you cutting in spending to match?