While the big shots in Washington politics were on vacation, the hardworking economists at Treasury released a technical study shining new light on the employment effects of allowing the top individual rates to return to 36 and 39.6 percent.
Using 2007 data, the study shows that between 35 and 47 percent of income from passthrough businesses (sole proprietorships, partnerships, and subchapter S corporations) would be subject to those higher rates. The new statistics -- whose earlier incarnations have been the subject of so much controversy -- are a welcome addition to the debate. (See Matthew Knittel, Susan Nelson, Jason DeBacker, John Kitchen, James Pearce, and Richard Prisinzano, "Methodology to Identify Small Businesses and Their Owners," Office of Tax Analysis, Technical Paper 4, Aug. 2011.)
But they are only part of the story. The new Treasury data can also be used to evaluate how effective high-end rate cuts are at targeting tax benefits to job creators. Of all the income affected by changing the top two rates, only about one-fifth is passthrough employer business income. In other words, 80 percent of income benefiting from the rate cut has no direct impact on employers owning passthrough businesses. Keeping tax rates low for business owners who have lots of other income is not an efficient way of getting those owners to create jobs. There are far more cost-effective approaches.
This raises the question: What are the economic benefits -- besides the positive effect on passthrough business employers -- of keeping high-end rates low? By asking this, we move the debate about extending the Bush tax cuts beyond the misplaced focus on small business to more familiar territory: How would preventing the return of the top rates to 36 and 39.6 percent affect the overall economy? From the perspective of Keynesian demand-side economics, keeping rates low would provide some stimulus, but it would be one of the least effective stimulus policies we could choose. As a tool for encouraging supply-side economic growth, it provides benefits, because a less progressive rate structure is always better for long-term growth. However, determining the size of these benefits -- like most issues in empirical economics -- is highly uncertain, and economists' conclusions always seem to align with their political affiliations. Whatever the size of the benefits, they must be weighed against the detrimental effects of a larger deficit (if in fact the increased revenue would be used for deficit reduction) and age-old concerns about tax fairness.
The Usual Back and Forth
In a nutshell, here is the debate so far. Ever since he was a candidate for the White House, President Obama has favored extending the Bush tax cuts for all Americans except individuals with more than $200,000 and families with more than $250,000 of adjusted gross income. Republicans have argued that this would have harmful effects on employment because much of the income of passthrough businesses is taxed at those top rates. For example, in a speech on August 22, 2010, House Speaker-to-be John A. Boehner, R-Ohio, informed his Cleveland audience: "According to an analysis by the nonpartisan Joint Tax Committee, Congress's official tax scorekeeper, half of small business income in America -- half -- would face higher taxes under the president's plan."
Democrats countered that argument by citing IRS statistics showing that high-bracket taxpayers make up a small percentage of owners of small business -- something in the neighborhood of between 1 and 4 percent, depending on how you measure it. For example, on August 24, 2010, Vice President Joe Biden disputed Boehner's claims: "He has created this myth that a tax cut for millionaires is actually a tax cut for small business. There aren't 3 percent of small businesses in America that would qualify for that tax cut."
Both Boehner and Biden are factually correct, but it is only the Biden statistic that is misleading. Wealthy people may own a small share of the number of passthrough businesses, but the size of the businesses they own are far larger than most. In The Wall Street Journal, Kevin Hassett and Alan Viard of the American Enterprise Institute set matters straight: "According to IRS data, fully 48% of the net income of sole proprietorships, partnerships, and S corporations reported on tax returns went to households with incomes above $200,000 in 2007. That's the number to look at, not the 3%" ("The Small Business Tax Hike and the 97% Fallacy," The Wall Street Journal, Sept. 3, 2010).
Now we can quibble about whether the percentage of passthrough income potentially exposed to higher rates is 45 or 48 or 50 or even 60 percent. (There is no one way to do the calculation.) But the point made by Hassett and Viard in their article stands irrespective of the computational details. Tax increases on high-bracket taxpayers will hurt passthrough businesses because a large share of all passthrough business activity is from businesses owned by those taxpayers. Democrats in and out of the Obama administration should stop trying to deny it. It cheapens the legitimate arguments against extending tax cuts for the wealthy.
New Treasury Data
Many commentators have rightly complained that a lot of passthrough income subject to high rates is not generated by entities that hire. So why should income from jobless businesses be included as part of that meritorious 48 percent Hassett and Viard are talking about? Some passthrough entities are effectively passive holding companies in which no entrepreneurial activity takes place. And even among those that are active businesses, many passthroughs do not hire any employees. In particular, a lot of sole proprietorships are independent contractors.
There is no perfect way to distinguish active from passive businesses, and there is no perfect way to distinguish potential employers from non-employers. In their study, Treasury economists developed quantitative measures to identify passthrough entities that are unlikely to ever employ anybody. With those entities removed, we can focus on the passthrough business employers that matter. To be considered a business in the Treasury study, an entity had to pass two tests: a de minimis activity test (total income or total deductions exceed $10,000, or their sum exceeds $15,000) and a business activity test (total deductions in excess of $5,000). And to be considered an employer, a business had to have labor deductions exceeding $10,000.
The impact of excluding nonbusinesses and non-employers is shown in Table 1. The first column of the table shows that income from active business employers was 68 percent of total passthrough income for taxpayers with more than $200,000 of AGI in 2007. Column 2 uses a different category of upper-income taxpayers but reaches roughly the same conclusion. Employer business income was 71 percent of total passthrough income for taxpayers in the 33 and 35 percent brackets in 2007.
Table 1. Eliminating Nonbusiness and Non-Employer Income
From Total Passthrough Income
33 and 35
|(1) Flow-through income||
|(2) Business flow-through income||
|(3) Employer flow-through income||
|Line (3) divided by line (1)||
|Source: Tables 13, 14, 17, and 18 of the August 2011 Treasury study. Income in the first data column includes business losses. The second data column includes only positive business income.|
The Treasury adjustments could still include a lot of high-profit businesses with relatively low levels of employment. Treasury business employer passthroughs could still include a hedge fund generating hundreds of millions of dollars in income but employing only a few people. Should we give the hedge fund managers millions of dollars of tax relief on the possibility that they might hire another office assistant? But as we said, no measure can be perfect. Treasury has done a reasonable and fair job of drawing a line that is inherently arbitrary.
A More Informative Statistic
As mentioned above, most of the technical debate about the effect of high-income rate increases on passthrough business has centered on whether the percentage of passthroughs (the Democrats' view) or the percentage of passthrough income (the Republican view) is the better metric of the impact of the proposed rate hikes. Despite tortured arguments from Democrats, clearly the second is superior. But even the second statistic is lacking as a meaningful guide to the debate. In particular, it gives us no information about the amount of business activity (and therefore the potential job creation) relative to the size of the tax change under consideration.
Table 2. High-Bracket Employer Passthrough Income as a Percentage of Total Passthrough Income and as a Percentage of Total High-Bracket Income Taxed at Regular Rates, 2007
(dollars in billions)
Republican political strategy is to shout down high-end rate increases because they would hurt the profitability of passthrough businesses. But at what cost are jobs saved? Preventing rate increases in the two highest brackets is a highly inefficient method of creating jobs. As the data below will show, only about one-fifth of income affected by high-end rate changes is from passthrough employer businesses. The other 80 percent of the tax cost has no direct effect on passthrough employment.
A tax benefit limited to profits of passthrough employers or, even better, to employment by passthrough businesses would be many times more effective at the same revenue cost. The failure of rate cuts to deliver tax benefits to small business has received scant attention. (For an exception, see Jane G. Gravelle, "Small Business and the Expiration of the 2001 Tax Rate Reductions: Economic Issues," Congressional Research Service (Jan. 6, 2011), Doc 2011-548 or 2011 TNT 7-28 .) Focusing on this lack of efficiency should be the starting point of the administration's defense of allowing tax cuts for the wealthy to expire.
Inefficient Employment Incentive
The left side of Figure 1 shows business income of high-bracket taxpayers as a percentage of all business income. High-bracket taxpayers receive 52 percent of all passthrough employer income. Although there are many differences in technical details, this 52 percent figure using new Treasury data is conceptually similar to the 48 percent figure cited by Hassett and Viard. This number tells us nothing about the efficiency of rate cuts for job creation at passthroughs.
For that information we need to look at the right side of Figure 1 and the bottom of the first column of Table 2. They show business income of high-bracket taxpayers as a percentage of all high-bracket income. (High-bracket income here excludes capital gain and dividend income subject to preferential rates.) Only 17 percent of the income of wealthy taxpayers is from passthrough employer businesses.
In Figure 1, high-bracket taxpayers are defined as taxpayers with more than $200,000 of AGI. This conforms with how IRS published data usually define income classes, and it is common practice for analyses of the distribution of business income to use this classification. But many taxpayers in this category are not subject to the 33 and 35 percent brackets. There are two reasons for this. First, families between $200,000 and $250,000 of AGI are below the threshold for the 33 percent rate. And second, a large portion of high-income taxpayers are subject to the alternative minimum tax with a marginal tax rate of 28 percent.
Figure 1. Passthrough Employer Net Business Income for Owners With AGI More Than $200,000
(Using New Treasury Data Set for 2007)
The new Treasury study also presents the distribution of business income by tax rate. As shown in column 3 of Table 2, taxpayers subject to the 33 and 35 percent rates, which are potentially subject to higher rates under the president's proposal, generated $249 billion of passthrough employer income. That's 35 percent of all passthrough employer income and 27 percent of AGI for taxpayers in those brackets.
Now, for completeness, we must get deep into details about the data. If Obama succeeds in allowing the highest rates to return to 36 and 39.6 percent, fewer high-bracket taxpayers will be paying the AMT. In other words, a larger proportion of taxpayers will be subject to higher regular rates than were subject to the 33 and 35 percent rates in 2007. Column 5 of Table 2 assumes that half the taxpayers in the 28 percent AMT bracket in 2007 will no longer pay AMT and instead will be subject to the 36 and 39.6 percent rates. Under this assumption, $292 billion of passthrough employer income accrues to owners subject to the reinstated 36 and 39.6 percent rates. As shown in Figure 2, that's 41 percent of all passthrough employer income and 21 percent of AGI for taxpayers in those brackets.
The refined Treasury data used in Figure 2 do not change the basic conclusion. Most of the tax effect of the high-end rate change has no direct impact on passthrough employers. The gross inefficiency of high-end rate reduction as a job creation policy provides the administration with an opportunity. As an alternative to the Republican proposal to extend rate cuts for all income tax payers, it can propose almost any sort of tax relief tied to wages paid. Few economists would be able to deny that an employment tax credit is far better targeted, and therefore a far more effective job creation mechanism, than high-end rate cuts.
Other Effects of Changing Top Rates
We can sidestep the whole overblown issue of small business's alleged special place in the economy. That's because even if there were an economic justification for providing tax breaks for small business employment, the discussion above shows that adjusting high-end rates is a poor method of delivering those tax benefits. However, before leaving the topic we should share two quick facts with readers. First, almost half of passthrough business income is not from small business (using the $10 million-of-receipts threshold suggested in the Treasury study). Second, most economists do not believe that small businesses generate more employment than large businesses. (See, for example, the Hassett and Viard article cited above; and John Haltiwanger, Ron Jarmin, and Javier Miranda, "Who Creates Jobs? Small vs. Large vs. Young," National Bureau of Economic Research Working Paper 16300, Aug. 2010.)
Figure 2. Passthrough Employer Positive Business Income for
Owners With AGI Potentially Subject to Higher Rates
(Using New Treasury Data Set for 2007)
So let's break out of the cramped confines of the political debate as it has been structured so far. The issue of the rate change effects on the overall economy is far more important than the effect of rate changes on small business. As with any major change in tax policy, we need to look at the economic impacts from both the demand and the supply side.
The most important goal of Keynesian economic policy is to stimulate aggregate demand when the economy is in recession. This can be done by increasing government spending or cutting taxes. For tax cuts to be effective, taxpayers must use their tax benefits to buy more goods and services. Of all possible types of tax cuts, those for the wealthy are least effective because the rich are more likely to save than spend. Keynesian stimulus is a policy focused on increasing employment in the short run. A rate cut for the wealthy is probably the last place politicians who are clamoring for rapid job creation should be looking.
Supply-side economics provides a far more flattering view of high-end rate cuts. Lower tax rates provide individuals more incentive to work, save, and invest. Because of the enormous political implications involved, the size of these effects has always been and always will be the subject of dispute. But we can say definitively that whatever the size of the effects, it is better for long-term economic growth to provide rate cuts for high-income households than low-income households. High-end rate cuts make tax rates more equal, and equalizing tax rates always provides a more efficient allocation of resources. Moreover, for the same reason that high-end rate cuts are bad from a demand-side perspective, they are good from a supply-side perspective: Rich people are more likely to save. And saving is the key to long-term productivity growth and competitiveness.
But that's not the end of the supply-side story. Now more than ever we must keep in mind the positive supply-side effects of deficit reduction. Tax increases on the wealthy or on anybody, if they are used to reduce the deficit, increase public saving. Supply-side growth is fueled by saving irrespective of whether it is from public or private sources. In a 2003 study mandated by a Republican-controlled House of Representatives, the JCT reported that the positive supply-side growth effects of rate cuts were offset by the negative supply-side effects of an enlarged deficit. (See "Mandated JCT Report Says House Bill May Hurt Economy," Tax Notes, May 19, 2003, p. 948, Doc 2003-12563 , or 2003 TNT 98-7 ; and "Macroeconomic Analysis of H.R. 2, The 'Jobs and Growth Reconciliation Tax Act of 2003,' Prepared by the Staff of the Joint Committee on Taxation," Congressional Record, House of Representatives (May 8, 2003), Doc 2003-11771 or 2003 TNT 91-83 .)
More on Small Business
Beside the facts presented so far, the Treasury report sheds a lot of new light on the size distribution of passthrough businesses. As we have already noted, the study shows us that a lot of passthrough businesses are not small. But perhaps the most striking revelation in the new Treasury data is that the ownership of these larger passthroughs -- particularly subchapter S corporations -- is highly concentrated among high-income taxpayers. If the goal is to help small business, a high-end rate cut is surely way off target. But that's enough for now. We will share more results from the new Treasury study on the smallness of passthrough business in our next article.