Reader Request: Do Lower Tax Rates Lead to Higher Tax Revenue?

A reader asks: If low tax rates lower income to the Treasury and cause deficits and lower economic growth, how do you explain how we ran deficits with a 70 percent top marginal tax rate in the 1970s and we ran surpluses for 1998-2001 with a 39 percent top marginal tax rate with almost identical average GDP growth for the periods? Doesn’t this fact give significant credence to the supply-side argument that lower tax rates increase tax revenue and cause surpluses?

Professor Chandra Mishra made roughly the same argument in our debate over the Bush tax cuts. I didn’t address it in my op-ed because I didn’t expect a tenured professor to advocate such a widely discredited position.

First, a clarification: I never said that “low tax rates…cause…lower economic growth.” On the contrary, the economic evidence indicates that tax cuts have a slightly positive effect in the short run.

In order for tax cuts to increase tax revenue, however, they would have to have such a large effect on economic growth that it outweighs the effect of the lower rates. Taking a smaller percent of a bigger number can yield more than taking a bigger percent of a smaller number, given the right numbers. At a certain point, if you keep raising taxes, people will stop working because it isn’t worth the effort. If enough people stop working, economic output decreases, and tax revenue shrinks despite higher rates. If you like graphs, you can visualize that “tipping point” as the top of the “Laffer curve,” named after economist Arthur Laffer who helped popularize the concept in the 1970s:  

The trick is figuring out the tipping point (t* in the graph). How high is too high?

Economists can estimate this tipping point by measuring “labor supply elasticity” — basically, the change in hours worked after historical changes in tax rates. According to three of the world’s top tax economists, “the best available estimates range from 0.12 to 0.40.” Using “the most reasonable” estimate of 0.25, the revenue-maximizing tax rate (t*) for the top income bracket is 69%.

Another recent paper used data from Australia, Canada, New Zealand, the UK, and the US. They found that the revenue-maximizing top tax rate ranges from 63% to 83%.

Therefore, the current top tax rate of 35% is well below the point at which tax cuts increase revenue.

This is no surprise to anyone with a decent recollection of recent history:

After [the 1981 tax cut], the deficit ballooned to 6 percent of GDP by 1983.

In the seven years [after the 1993 tax increase], the country went from a federal deficit of 3.9 percent of GDP to a surplus of 1.4 percent.

In the year the 2001 tax [cut] was adopted, there was a surplus of 1.3 percent of GDP. This turned into a deficit of 3.6 percent by 2004…

The national debt has followed a similar pattern, rising by an astounding 14.8 percentage points relative to GDP over the 7 years following adoption of the 1981 supply-side tax cuts, shrinking by almost 10 percentage points relative to GDP following 1993, and moving back up by 3.8 percentage points relative to GDP after the 2001 tax cuts.

Again, if you prefer graphs, here’s one from the Center for American Progress:

And one from the Center on Budget and Policy Priorities:

So, what explains the paradox in the reader’s question?

First, his numbers are wrong. We did not have “almost identical average GDP growth” in the 1970s and 1998-2001. From 1969 to 1979, GDP grew from $984 billion to $2.562 trillion, for an annual growth rate of 10.0 percent. From 1997 to 2001, GDP grew from $8.332 trillion to $10.286 trillion, for an annual growth rate of 5.4 percent.

So, it’s completely untrue that a lower top tax rate resulted in higher GDP growth, thus falsifying the supply-siders’ entire argument. This is unsurprising, however, given the overwhelming evidence that tax cuts didn’t spur superior economic growth.

Second, the periods aren’t comparable. The 1970s included a deep recession that lasted 16 months. 1998-2001 was the climax of an unsustainable stock market bubble.

Third, from 1993 to 2001, tax revenue increased significantly as a percent of GDP, further proving that the cause was not an increase in hours worked or economic output. At the same time, federal spending decreased, as a percent of GDP, to the lowest level in over thirty years — significantly lower than it averaged during the 1970s:

Finally, the reader’s comparison ignores all the years when we had high deficits with low rates and low deficits with high rates. It’s cherry-picking, it’s unscientific, and it’s incredibly misleading. Paul Krugman said it best:

If the tax-cut movement attributed the booming economy of 1999 to a tax cut Reagan pushed through 18 years earlier, why didn’t they attribute the economic boom of 1983 and 1984 — Reagan’s ”morning in America” — to whatever Lyndon Johnson was doing in 1965 and 1966?

By 1998, the top tax rate had been 39.6% or lower for 11 years. For most of those 11 years, the budget deficit (as a percent of GDP) was the same as or bigger than it had been throughout the 1970s:

Tax cuts for the rich in the 1980s didn’t suddenly trigger a surge in tax revenue in 1998. Instead, the budget surplus resulted from the stock market bubble, the decline in federal spending, and the tax increase of 1993.