Tax Plans: Some Useful Graphs

On Wednesday, I posted about an online article (which I first saw at Greg Mankiw's blog) by Alex Brill and Alan Viard in the AEI's "The American" titled The Folly of Obama's Tax Plan. I argued and still believe that the way that article is written is deeply misleading (one of the authors responded in comments, and my point-by-point response is here; but enough food-fighting -- this is a substantive post!).

My main point was that

An honest article meant to inform would show both the marginal tax rates graph and a graph of after-tax-and-transfer income against pre-tax-and-transfer income. I assure you that there is a very good reason that Brill and Viard don't include that second graph.

The original Brill and Viard piece was a vehicle to publicize a chart showing how the Obama plan would increase effective marginal tax rates for a carefully chosen type of family, at all income levels above $25k (for some reason, Brill and Viard's chart doesn't show effective marginal tax rates for those with pre-tax income below $25k). In isolation, this graph has the potential to confuse readers unfamiliar with the distinction between "taxes" and "effective marginal tax rates". As I suggested in the text I quoted just above, a fair-minded way of discussing the Obama plan's effects--even a discussion intended to raise the question of disincentive effects--would not focus only on effective marginal tax rates. Rather, such a discussion would show how Obama's plan would affect after-tax-and-transfer income for people with different levels of pre-tax income. Unless I've missed it, Brill and Viard have yet to provide such a graph or discussion.

In a happy coincidence, Obama economic advisers Jason Furman and Austan Goolsbee published an op-ed in the WSJ concerning Obama's tax plan. This op-ed led the Tax Policy Center (original source of the data used by Brill and Viard) to re-run their numbers. You can see some summary results of TPC's simulations of the Obama plan here; these results incorporate the details of the Furman-Goolsbee article. For comparison's sake, you can also see results from the same types of simulations applied to McCain's tax plan (dated July 23, 2008, and based on descriptions by McCain advisers); these results are here.

Since the TPC numbers break down the effects of each candidate's plan on various pre-tax income categories, I thought it would be helpful to provide the sorts of graphs that Brill and Viard didn't. Each graph shows the effect of Obama's plan side-by-side with the effect of McCain's plan (the analyses are relative to what is called the "tax cuts extended baseline", which I believe means that the analysis assumes extension of the Bush tax cuts; this isn't my preferred analytical baseline, but some people like it, and showing both plans relative to the same baseline removes any relative distortions).

Three notes before I get started:

1. I haven't seen similar TPC-done effective marginal tax rate tables for the two plans (I searched the site but did not find them). If someone can point me to them, I'd be delighted to make similar graphs for effective marginal tax rates in a future post.
2. The point of this post is to provide interested readers with information helpful to understanding what each proposal would do. Obviously I'm partial to Obama (look at the blog's title!), but I've tried to keep editorializing to a minimum in this post. Look at the facts, and make up your own mind.
3. The TPC does its estimates under various assumptions about the plans and the tax-law/tax-policy baseline. I made these tables using the links provided by Pete Davis of the Capital Gains and Games blog, which is where I first saw that the TPC had updated its Obama-plan analysis. Analyses based on other plausible assumptions about the plans and/or future counterfactual tax rates would of course yield different charts. My guess, though, is that the basic story would be the same. I'm of course happy to be shown that I'm wrong on this guess. [Update: Turns out that the TPC analysis of the McCain plan based on McCain's stump speech is very different from the analysis I discuss here, which is based on what his economic advisers say. In short, the McCain stump-speech version would lead to considerably larger reductions in taxes paid by most groups than the version based on what his economic advisers told TPC. The overall cost of the stump-speech plan is $2.8 trillion more than the economic-advisers one. To see this, compare the next-to-bottom-row figure in the far-right column of this table (analysis of stump-speech version) to the same figure in this one (analysis of economic advisers' version). The same comparison between Obama's stump and advisers' versions involves only a $400 billion difference in his plan's total cost (the stump version is actually less expensive). I'm going to leave the rest of this post as-is, focusing on what advisers told TPC. When I have some time later, I'll try to put together a side-by-side of McCain's stump versus Obama's, as well as McCain's advisers versus McCain's speeches.]
   

The first pair of charts concerns the percentage change in after-tax income that would result from each plan (click on the image to see it in larger form).


As you can see, the Obama plan leads to a large percentage increase -- 8.6% -- in after-tax income for households with pre-tax income below $10k. In fact, Obama's plan increases after-tax income for those with pre-tax incomes in every category below $200k. McCain's plan would increase after-tax income for every category, though the only categories for which after-tax income would rise by more than 1% are those with pre-tax income between $500k-$1mil and $1mil+. By contrast, Obama's plan would reduce after-tax income by 1.5% for those making $200k-$500k, by 5.6% for those making $500k-$1mil, and by 11.3% for those making $1mil+.

Now let's consider actual dollars. My next charts shows the change in the net tax bill that each plan would cause for the different income categories. Note that a negative number is a tax cut, while a positive number is a tax increase. Chart A shows the change under each plan for all categories with incomes below $200k; I discuss those with higher incomes in Chart B, below.

This chart shows that both plans would reduce the net taxes paid for all groups with incomes below $200k. But Obama's plan cuts the net tax bill considerably more than does McCain's for those making below $100k. Here are the numbers by income category:

1. Pre-tax income less than $10k: Obama tax cut is $487, McCain's is $14.
2. Pre-tax income $10-20k: Obama tax cut is $709, McCain's is $26.
   
3. Pre-tax income $20-30k: Obama tax cut is $896, McCain's is $89.
4. Pre-tax income $30-40k: Obama tax cut is $1,039, McCain's is $178.
   
5. Pre-tax income $40-50k: Obama tax cut is $1,124, McCain's is $232.
   
6. Pre-tax income $50-75k: Obama tax cut is $952, McCain's is $336.
   
7. Pre-tax income $75-100k: Obama tax cut is $779, McCain's is $483.
   
8. Pre-tax income $100-200k: Obama tax cut is $407, McCain's is $847.
   

On average, McCain's plan cuts the net tax bill by $601, while Obama's increases it by $549. Thus, McCain's plan further increases federal debt over the tax cut extended baseline, while Obama's plan reduces it (note that both plans increase federal debt over current law, under which all Bush tax cuts are set to expire; McCain's increases federal debt by more than Obama's).

Chart B shows the change in net tax bills for the top three pre-tax income categories. Notice that the vertical-axis scale of this chart is very different from the scale in Chart A (that's why I broke the categories into separate charts).
Here are the numbers:

1. Pre-tax $200-500k: Obama raises taxes $3,546, McCain cuts $1,892.
2. Pre-tax $500-1mil: Obama raises taxes $30,499, McCain cuts $6,825.
3. Pre-tax $1mil+: Obama raises taxes $262,371, McCain cuts $58,632.

In sum, Obama's plan involves

• large increases in taxes paid by the highest-earning Americans, who represent a very small share of people but take in a (relatively) very large share of pre-tax income, coupled with
  
• moderate-to-large reductions in the net tax bill faced by everyone else.

By contrast, McCain's plan involves

• smaller, but still large, reductions in taxes paid by the highest-earning Americans, coupled with
  
• very small reductions in the net tax bill faced by everyone else.

(I stress again that both plans add substantially to the federal debt relative to current law, which means that someone will have to pay the bill later; again, though, that bill would be smaller with Obama's plan than with McCain's.)

I now turn to the issue of average tax rates. Whereas the effective marginal tax rate answers the question, "What share of the next dollar that I earn will I have to pay in taxes?", the average tax rate answers the question, "What share of all my income do I have to pay in taxes?" Because marginal tax rates rise with pre-tax income in the U.S. tax system, and because of the way various tax credits are phased out as pre-tax income rises, effective marginal tax rates can be very different from average tax rates. The question of whether marginal tax rates or average tax rates are more important doesn't make a lot of sense, because they're both relevant to answering different questions. A person deciding whether to work one more hour should be more concerned with marginal tax rates, while a person deciding whether to work at all, or between part- and full-time work, should be more concerned with average tax rates. A person concerned with distributional fairness (whatever said person thinks that means) should probably be more concerned with average tax rates, though the disincentive effects of marginal tax rates do impact the efficiency costs of a given degree of redistribution.

With that said, my final chart shows the change in the average tax rate paid by people in each category. A negative number means that the average tax rates paid by those in the group would fall under the plan (a tax cut), while a positive number means the average tax rates would rise (a tax increase). The units of the chart are in percentage points, not percent. That is, the -8.1 figure for those in the lowest income group under Obama's plan means that instead of paying 5.3 percent of their pre-tax income in taxes, on average they would receive a 2.8 percent rebate (5.3 percent minus 2.8 percent equals 8.1 percentage points).


This chart tells a pretty simple story. McCain's plan would reduce all groups' average tax rates, though generally by very little. Except for those making more than a million dollars a year, who would see a drop in average tax rates of 1.8 points, McCain's plan would reduce the average tax rate by less than one percentage point across the board.

By contrast, Obama's plan would reduce average tax rates by a moderate to large amount (between 1.4 and 8.1 percentage points) for all groups with pre-tax income below $75k; those in the $75-100k and $100-200k groups would see average tax rates fall by 0.8 and 0.3 percent. Most notably, those in the three highest groups would see moderately small (1.2 percent for $200-500k) to large (4.1 and 8.0 percent for those making $500k-$1mil and $1mil+) increases in their average tax rates.

Update: Don Pedro has more on other aspects of the WSJ op-ed piece.