At the first presidential match-up in Denver last week, Mitt Romney injected new charges into the discussion of what will happen to small businesses if taxes are raised on the wealthy. Speaking of President Obama's proposal to let the Bush-era tax cuts expire for those who earn more than $250, 000, the Republican candidate told the president that the impact would be felt by only 3 percent of businesses but that those businesses "happen to employ half - half - of all the people who work in small business. Those are the businesses that employ one-quarter of all the workers in America. And your plan is to take their tax rate from 35 percent to 40 percent. The National Federation of Independent Businesses has said that will cost 700, 000 jobs."
Startling numbers, but are they accurate? Well, the first two are not, and the last is disputed.
Let's look at the share of employment first, Mr. Romney's charge that letting the Bush tax cuts expire for the top two tax brackets will specifically hit businesses that employ half of all the people who work in small businesses and a quarter of all American workers. The Romney campaign declined to discuss its source, but the assertion closely echoes a figure circulated by the N.F.I.B., when the advocacy group sent a letter to Congress two years ago opposing any tax increases, including on the wealthy.
That letter said, in part: "Based on an N.F.I.B. small-business survey, the businesses most likely to face a tax increase by raising the top two rates are businesses employing between 20 and 250 employees. According to U.S. Census data, businesses with between 20 and 299 workers employ more than 25 percent of the total work force."
There are two things to say about this statement, as retold by Mr. Romney. First, there is the obvious error of transmission in his retelling — that these companies account for a quarter of the work force, when in fact the Census Bureau cohort included companies with up to 299 employees (not 250). But the bigger issue is the implication that all of these businesses (by Mr. Romney in the debate) or most of them (by the N.F.I.B. in its letter) face a tax increase under the Obama plan.
The $250, 000 threshold
In fact, the survey (pdf) at the root of the claim suggests the opposite — that a minority of these businesses in this size category, about 22 percent of them, earned more than $250, 000 in business income, the threshold for higher taxes under the Obama plan. According to the Census Bureau figures that the N.F.I.B. relied on (available from the S.B.A.), in 2007, when the N.F.I.B. poll was taken, businesses with 20 to 299 employees employed almost 28 percent of the total work force. Assuming that the N.F.I.B. survey results can be extrapolated to the broader small-business population*, 22 percent of 28 percent is 6 percent. That means that, according to the N.F.I.B. survey, the share of the total work force that might be affected by the increase in tax rates is around 6 percent, not 25 percent.
(Interestingly, in an interview for this post, an N.F.I.B. tax lawyer, Chris Whitcomb, seemed to back away from the claim that most of the small businesses likely to face the tax cut are those that employ 20 to 250 workers. Referring to the passage in the N.F.I.B.'s letter to Congress that cited the survey, Mr. Whitcomb said, "All that statement is saying is that looking at this survey data, of businesses that have 20 to 249 employees, 22 percent of those have a business income above $250, 000. The conclusion that we're drawing from that is that this group will be disproportionately impacted by raising those rates.")
As for Mr. Romney's suggestion that the tax increase will affect companies employing half of the people who work at small businesses, that also is not accurate. In 2007, according to the Census data, companies with 20 to 299 employees employed 55 percent of the total work force at companies with fewer than 500 workers, a common definition for a small business. But given the N.F.I.B. survey's finding that only 22 percent of businesses in roughly that size range would actually face a tax increase, those businesses' share of the small-business work force would be only about 12 percent.
Mr. Romney's larger point is that the 4 percent increase in taxes on small businesses making more than $250, 000 a year will cause these profitable companies to cut jobs - 700, 000 employees in all, he said. For this assertion, Mr.Romney again relies on the N.F.I.B., and a recent analysis by the accounting firm Ernst & Young, commissioned jointly by the Independent Community Bankers of America, the N.F.I.B., the S Corporation Association, and the United States Chamber of Commerce. The analysis does indeed claim that raising taxes on those making more than $250, 000 will cost the economy 710, 000 jobs.
Effect of taxes on hiring
But the effect of taxes on investment (and, hence, on hiring) is hotly contested along ideological lines. (The Agenda discussed this earlier this year when the Buffett Rule was the subject of conversation.) Liberal economists see little, if any, effect from taxes on investment. Conservatives see a substantial effect, but even some conservative economists acknowledge that it is impossible to prove the connection using economic data given all of the other variables at work in the economy. Instead, they sometimes make their case on theoretical, or even visceral, terms. Or they publish analyses like the one from Ernst & Young, which is not an empirical study, using real-world data, but one making predictions based on the authors' assumptions about the relationship between taxes and investment. The authors of the Ernst & Young report are established conservative economists and the paper relies heavily on the work of other Republican-affiliated economists.
Not surprisingly, liberal economists dismiss it. The study's authors "get a considerably larger employment response from high-income tax cuts than other standard models or, more importantly, historical experience, " Jared Bernstein, a former adviser to Vice President Joseph Biden, told The Agenda. (Mr. Bernstein also criticized the report in an August blog post.)
That critique won't find a sympathetic ear in the Romney campaign. After the debate, Paul Ryan, Mr. Romney's running mate, mentioned the 700, 000 jobs that could be expected to be lost in a speech in Virginia. And it seems likely to come up again in the vice presidential debate Thursday night.
*The N.F.I.B. survey, conducted by Gallup in December 2007 and January 2008, interviewed only 154 businesses with 20 to 249 employees, which statisticians would consider a small sample. The businesses in the survey were randomly selected from Dun & Bradstreet lists.