Tax Analysts®Tax Analysts®

Article Archive

September 22, 2011
News Analysis: Buffett Bracket or Roosevelt Rule? Two Ways to Skin a Fat Cat
Joseph J. Thorndike

Full Text Published by Tax AnalystsTM

What's the best way to extract more revenue from the rich? When President Obama formulated his "Buffett rule," he avoided that question. But it's an important one, in terms of both substance and symbolism.

In his deficit reduction plan, Obama provided the official formulation of the Buffett rule, named for famed Omaha, Neb., investor Warren Buffett: "No household making over $1 million annually should pay a smaller share of its income in taxes than middle-class families pay."

That's a very clear statement with very murky policy implications. In fact, it's more of a guideline than a rule, and even as a guideline, it offers precious little guidance. Implementation of the Buffett rule could take any of several forms -- and probably more than one. (For related coverage, see p. 1317.)

Up to Congress

When asked for specifics, Treasury Secretary Timothy Geithner said, "We're not going to give the Congress a detailed proposal for how to meet that principle, because we think there are a bunch of different ways to do that."

Well, that's convenient. But it hasn't stopped the cries of outrage coming from conservatives. Absent the details, it's hard to know what right-leaning critics dislike so much. After all, at least some techniques for implementing the rule have Republican support. Or at least they once did.

For instance, one plausible modality for the Buffett rule might be a cap on itemized deductions, limiting them to 28 percent. The Obama administration has in fact proposed that rule. But so have various right-leaning economists, including Martin Feldstein ( and R. Glenn Hubbard (

A deduction cap would certainly run afoul of antitax hawks in the Republican Party, including Grover Norquist of Americans for Tax Reform. But the idea still has a plausible claim to bipartisan support.

An Alternative AMT

That probably can't be said for another likely means of implementing the Buffett rule: a new version of the alternative minimum tax (presumably, although not necessarily, one operating alongside the existing AMT). The very notion of that beast is hard to contemplate. While the AMT probably serves its intended purpose (at least part of the time), it's also broadly reviled for its complexity -- not to mention its stealthy approach to raising taxes on middle-income individuals.

A third possibility for implementing the Buffett rule is the most straightforward: simply raise rates. That approach would presumably involve higher rates for both rich people and their sources of income. After all, raising rates on regular income won't do much to solve the Buffett problem as long as capital gains are taxed at highly preferential levels.

It's safe to say, however, that rate increases of any sort would draw the ire of many economists and pretty much all Republicans. Still, they have a certain symbolic virtue, clarifying the distributional issues at stake in this debate over fairness and fiscal policy. And they have plenty of historical precedent.

The Roosevelt Rule

When it comes to issues of progressive fairness, the obvious place to look for left-leaning outliers is the New Deal. In 1935, at the behest of President Franklin D. Roosevelt, Congress raised income tax rates almost across the board. In particular, lawmakers focused on creating new brackets for the superrich, persuaded by FDR's argument that further differentiation at the top was morally vital.

Roosevelt made his case in a special message to Congress on June 19, 1935:

    The application of the principle of a graduated tax now stops at $1,000,000 of annual income. In other words, while the rate for a man with a $6,000 income is double the rate for one with a $4,000 income, a man having a $5,000,000 annual income pays at the same rate as one whose income is $1,000,000.

To Roosevelt, there was a meaningful difference between a man with $1 million in annual income and someone with five times that amount. Congress agreed and established new brackets beginning at $2 million and $5 million. Adjusted for inflation, those brackets would start at something like $32 million and $80 million in annual income today.

Rates in the new brackets were 78 percent and 79 percent, respectively. The old $1 million bracket was set at 77 percent. The new brackets, in other words, didn't provide a lot of extra graduation. But then they didn't have as much room to rise as we have today.

Issues of high-end progressivity were personal in the 1930s, much like they are today. If Buffett is the poster boy for spare-the-rich taxes in the early 21st century, then other business titans played a similar role in Roosevelt's day. In fact, the top bracket enacted in 1935 applied to precisely one taxpayer: John D. Rockefeller Jr.

The Decline of Superbrackets

Superbrackets fell out of favor during World War II. With rates rising dramatically for everyone, lawmakers chose to lower the threshold for entry to the top brackets, chiefly as a way to raise more money. Starting in 1942, the top bracket began at $200,000 and featured a rate of 88 percent.

Lawmakers never returned to the land of superbrackets, but the rate structure still differentiated between the simply rich and the very rich. In 1956, for instance, upper-echelon brackets began at $200,000, $300,000, and $400,000 (roughly $1.6 million, $2.5 million, and $3.3 million today, according to figures from the Tax Foundation). Rates in these brackets were 89, 90, and 91 percent.

By comparison, our current bracket system is stunted. The top bracket, beginning at $379,150, is roughly equivalent (in inflation-adjusted terms) to the 1956 bracket that began at $44,000. That bracket fell squarely in the middle of the rate structure, with 11 brackets below and 12 above.

Of course, the closer you get to the present day, the less graduation there is at the top of the income spectrum. And for many good reasons. Excessively high rates breed avoidance, distort decisions, and generally raise all sorts of economic havoc.

But rates don't have to return to Eisenhower levels if all we want to do is introduce some additional graduation to the rate structure. With our current top rate of just 35 percent, there's plenty of room for making additional distinctions at the top end.

There are also plenty of reasons to do it. It's wrong to conflate the wealth of a millionaire with the wealth of a billionaire. If $375,000 is a lot of money, it's nothing compared with $3.75 million. Or $3.75 billion. And it's not like these numbers are fanciful.

Implementing the Buffett rule won't be easy. And it may not even be hugely successful. But the game is worth the candle, if only to drive home the importance of fiscal equity in an era of (inevitably) rising taxes.