## Cain’s Regression Analysis Fails to Reassure Me

For some reason, I guess because it’s getting some serious attention, I found myself wondering if Herman Cain’s 9-9-9 tax plan raises the same amount of revenue as the current federal tax system.  The WaPo says the following:

“Cain has said that his plan would collect about the same amount of revenue that now flows into the government, and that it would increase as the economy strengthened. Experts, however, say that is difficult to know because the candidate has given only broad outlines for his proposal…”

I checked with some colleagues who tell me that the usual scorekeepers of such things haven’t looked into this yet, probably because they need more than three numbers—or one number three times.

So what makes Mr. Cain so confident about revenue neutrality?  After all, he exempts capital gains, estates, the payroll tax, and the corporate part of dividends—that’s a lot of the base.

I went to his web site to learn more.  I found this point, under “Phase 1 – 9-9-9:”

“The Phase 1 Enhanced Plan incorporates the features of Phase One and gets us a step closer to Phase two”

…which helped a lot, but I wanted more.

Thankfully, a colleague sent me this, from a recent TV appearance by the candidate (hat tip, CCH):

“I had some of the best economists in this country help me to develop this plan. You know, my background is mathematics. It was a simple regression analysis. We took the government data and looked at how much tax revenue from personal income tax, how much tax revenue came from corporate tax, how much revenue came from capital gains tax, how much revenue from the death tax. We added them all up, and you do a simple regression analysis and say in order to reduce this much on corporate income, personal income and national sales tax, what should that number be if we equally break up those three buckets. It was a simple regression analysis.”

OK.  Your background is math…that’s good.  But regression analysis…really?  What are you talking about?  Can you just say freakin’ anything and get major freakin’ coverage in a major freakin’ paper?

Apparently so, but bad for the WaPo for its shoddy evaluation of this question—the framing above re revenue neutrality just says “Cain says it is; experts aren’t so sure.”  What it should say is: Cain was unable to provide any evidence to support his claim.

### 14 Responses to “Cain’s Regression Analysis Fails to Reassure Me”

1. Sandwichman says:

We should just be thankful it isn’t 6-6-6! You would have to be raising Cain about the number of the beast. How’s THAT for a regression analysis?

• Jared Bernstein says:

If you turn it upside down, it is 6-6-6…yikes!

• Sandwichman says:

Do you think maybe “turning it upside down” is what Cain means when he says “regression analysis”?

• Sandwichman says:

Does this make “2/3″ the Cainsian multiplier?

• sven says:

Sandwichman FTW!

2. Scott Supak says:

Evidence? He’s a Republican. He doesn’t need any stinking evidence!

3. perplexed says:

-”After all, he exempts capital gains, estates, the payroll tax, and the corporate part of dividends—that’s a lot of the base.”

How else are we to secure the families of the oligarchs in perpetuity?

What he meant to say was “regressive” analysis; remember, his background is in “math,” not English or communications.

4. Walt French says:

Y = â × 1.0

Least squares does a fine job of finding averages.

Goodness-of-fit stats, too!

5. joe says:

Don’t know what sales he is taxing but two of the 9s are very clear. He is taxing corporate profits at 9% and salaries and wages at 9%. So if you get corporate profits from the commerce department for 2008 and salaries and wages from the IRS for 2008, you have 1.24 trillion in corporate profits and 5.9 trillion in salaries and wages. Tax both at 9% and that’s 642 billion dollars of revenue. In 2008, tax revenue was 2.54 trillion. So 1.89 trillion in revenue would have to come from a national sales tax. Which means sales would need to be 21.08 trillion or over 150% of GDP which was 14.29 trillion in 2008.

• Jared Bernstein says:

Yeah…but he did regression analysis!

You’d need to back deductions and such out of these aggregates, but I think you’re heading in the right direction.

• sven says:

• ModeratePoli says:

@ joe, thanks for the simple math (the best kind of math). It really shames political reporters that they didn’t do this.

I’ve seen only one report on where the 9% came from:
http://www.suntimes.com/news/politics/7952138-418/hermain-cain-to-be-at-teacon-in-schaumburg-this-weekend.html

According to this article, Cain said that it was based on the tithing for churches, but the government should get less than the church, so he made it 9%. Incredible. This should put an end to his surge, but I have no faith that populist Republicans will see it that way.

6. Ken Houghton says:

Assume economic theory works perfectly. (Well, it’s Fantasyland.)

The 9% corporate tax and the 9% personal tax produces 9% of GDP between them. (Note I am ignoring depreciation as a deduction, but we can sidebar that.)

The national Sales Tax is 9%. It is assessed on purchases. At the maximum left–91% of GDP–this produces 0.91*0.09 = 8.19% of GDP

9% + 8.19% = 17.l9% GDP.

And, as noted, this is the BEST case scenario (no depreciation, no direct social safety net funding, salaries but nothing else deducted from corporate income).

You don’t need models, regression analysis, or even pre-calculus. Algebra, which is apparently beyond the abilities of the WaPo political staffers, works quite well.

The last time Federal spending was that low was basically the last year before Medicare became fully effective (1966). If you end Medicare (all parts) and Medicaid–and never have a recession, since there are several other years (’52, ’53, ’58, ’62, and ’63) that miss the maximum revenue Cain’s “plan” projects–you could almost make it work. With no investment.

I suspect you also have to declare all extant Federal debt null and void; otherwise, interest payments become an issue to more of an extent than they were in the Capped Yields era.

That’s just using FRED data (AFEXPND/GDP) from 1947 forward, so you can’t really have–well, pay for–any more wars either. (Yes, I know we’re not paying for the current ones. But dropping the pretense that’s not official policy?)

7. Craig says:

If adding up a list of numbers and then dividing it by three is “regression analysis,” then why the hell did I do so much work in my college statistics classes?