Earlier, my colleague Derek Thompson wrote about Paul Krugman's claim that ugly deficit projections would look far better if Bush had done things differently. I wanted to add a little to Derek's analysis. What I'd like to contribute is some skepticism.
Anyone who reads much Krugman knows that he
would sooner jump into Niagara Falls than say anything positive about President Bush was not President Bush's biggest fan.* But that doesn't mean bias is the only thing driving his claim that the deficit picture would look a lot better if Bush had done some things differently. I think that's probably right. The flaws I see in his argument stem more through his specific analysis, which I think is far from rigorous. Yet, I don't know if it's as much Krugman's methodology as it is the general problem in "what if" scenarios like this one.
First, let me quote the relevant passage from Krugman:
There were two big-ticket Bush policies. One was the tax cuts, which cost around $1.8 trillion in revenue; add in interest costs, and we're presumably talking about more than $2 trillion in debt. The other was the Iraq War, which has cost at least $700 billion, and will cost more before we finally extract ourselves.
Let me begin by saying that I was also not a fan of Bush's fiscal strategies. While I generally support tax cuts, in most cases I do not support them in the absence of spending cuts. Indeed, Bush was spending even more while cutting taxes, which is a terrible idea. I also prefer not to get into the Iraq War question, because I don't intend to dispute that part of Krugman's claim.
With that said, the kind of analysis Krugman is trying to do regarding the loss in tax revenue troubles me. I don't like his claim that, without the tax cuts, the country would have had $1.8 trillion more in revenue, because it cannot be accurately verified. Just because taxes were cut by that amount doesn't mean that the same amount was a pure loss for the government.
This $1.8 trillion made its way into the private sector, after escaping Uncle Sam's clutches. Then, it stimulated the economy. Some was directly invested; some was indirectly invested as consumer spending contributed to business profits resulting in greater investment. That, consequently, should have caused gross incomes to have increased more than they would have without the tax cuts.
Let me explain a fictional example of what I'm talking about. Imagine a person who was making $100,000 per year before the tax cuts, putting him in the 30 percent federal tax bracket. After the tax cuts, a few years later, his tax bracket changed to 28 percent. But he doesn't necessarily pay 2% less taxes -- his income should have increased more than it would otherwise due to better economic growth resulting from the tax cuts. If that effect was an additional $10,000 per year giving him a $110,000 income, then he's actually paying $30,800 in taxes. That's $800 more than the $30,000 he would have paid before the tax cut.
This example is fictional because it has to be. It's impossible to know how much incomes changed and how much tax revenue was made up. But even if the income of the person I made up increased only 3% instead of 10%, it's still just as false to say the government had 2% less tax revenue from him because his bracket changed from 30% to 28%.
Tax cuts also have a positive effect on employment levels. They allow businesses to hire more workers, reducing unemployment. With more people employed, more will pay taxes, increasing tax revenues for the government.
So am I claiming that Bush's tax cuts actually gave the government more tax revenue than it would have had beforehand? Absolutely not. That could be true, but I don't pretend to know if it is. I'm agnostic on the question, because it's too complex to be able to answer.
That's the problem with "what if" scenarios like this. I can't be sure the positive effect that the tax cuts had on employment and incomes. But if economics teaches us anything, then we know there was some positive effect. That's why Krugman's analysis is overly simplistic, and his methodology cannot help but produce a false estimate of the loss in tax revenue. You can't just assume tax cuts result in a gross loss of tax revenue identical to the portion of the cuts.
*Sorry -- didn't want snarkiness to get in the way of my argument.