EconomistMom.com
…because I’m an economist and a mom–that’s why!

EconomistMom.com

The Tax Policy and the Economy Fairy Tale

March 12th, 2012 . by economistmom

My latest Tax Notes column which came out today (subscription-only access here) is basically a recap of my testimony on March 1st before the Senate Budget Committee.  It’s a written version of the script I used for my oral remarks, plus a chance to report (or vent) about the line of questioning that came from one of the Republican senators that day.

About the basic premise of the hearing, which was called “”Tax Reform to Encourage Growth, Reduce the Deficit, and Promote Fairness,” I explain that:

I recently heard the three tax reform goals listed in the hearing’s title referred to as a “fiscal trilemma,” suggesting it might not be possible to achieve them all.2 Equating the three with a dilemma suggests that working toward them will be a negative experience. Indeed, many policymakers are caught speaking of at least one of the goals with partisan disdain: “encouraging growth” (a popular Republican goal) might be labeled by some Democrats as “pandering to the rich”; “reducing the deficit” by including at least some new revenue (a popular centrist goal) might be labeled by some on both sides as “killing jobs”; and “promoting fairness” (a popular Democratic goal) might be called “class warfare” by some Republicans.

Nonpartisan economists would respond that all three goals will benefit the economy. And the good news is that it really is possible to find tax policy changes that would help achieve all three goals — and possibly help achieve simplicity. That good news is doubled by the recognition that different policymakers actually like all the goals more than they’ll admit in public, but they assign different implicit weights to the different goals — suggesting that the only way to ensure bipartisan agreement is to make sure a proposal helps achieve all three goals.

I then explain the problem with the tax policy “fairy tale” that sounds so happy and easy:

[M]any so-called tax policy experts spin a simple fairy tale when they talk about how to reform the tax system. They say that we just need to cut tax rates, which will expand the economy, which in turn will reduce the deficit. But unfortunately, in the real world, we face real budget constraints and a real scarcity of resources. Real economists know that optimizing means not just maximizing benefits but weighing benefits against costs so that benefits net of costs are maximized. In the context of the real world and our experiences with the economic effects of different tax policies, cutting tax rates to achieve all of our goals is pure fantasy.

I made three main points in my testimony regarding the goals of encouraging growth, reducing the deficit, and promoting fairness:

    1. It is impossible to expand the supply side of the economy through continued, seemingly easy, deficit-financed tax cuts.2. It is impossible to reduce the deficit without allowing, and even seeking, higher revenues as a share of our economy.

    3. It is impossible to promote fairness in the tax system without raising tax burdens on the rich.

And the only part of the Q and A where I have to admit I felt a bit “bullied” was this:

One of the more hostile exchanges at the hearing was when Sen. Ron Johnson, R-Wis., questioned what we thought the maximum marginal tax rate should be. Each time [Len] Burman [of Syracuse, the other witness invited by Chairman Conrad] and I tried to respond that it depends on the breadth of and distortions within the existing tax base, Johnson interrupted and insisted on our providing a specific number without any qualifications. It was obviously a setup, as [Dan] Mitchell [the Republican witness from the Cato Institute] described in a blog post. Although I reluctantly gave a specific answer of 70 to 80 percent, I wasn’t advocating a marginal tax rate that high but only responding that a total marginal tax rate — combining taxes at all levels of government — any higher than that would be a bad idea. I tried to point out that the maximum marginal tax rate could mean the rate on the richest person in the country’s last dollar earned. I believe 70 to 80 percent is around Laffer curve levels — the highest rate possible before revenue is lost. [In the Tax Notes column I cited this NBER paper by Christina and David Romer.]

That maximum marginal tax rate is totally different, however, from the survey results Mitchell cites that show people not wanting anyone to be taxed at more than 30 percent. Mitchell understandably likes the interpretation that ordinary Americans are referring to the maximum top marginal tax rate bracket. But I really doubt that most Americans understand the difference between marginal and average tax rates, or if they do, that they are inclined to automatically think top marginal rate (on the last dollar earned) when asked about the maximum tax rate that top earners should pay. When thinking about what’s fair, I think most people have in mind the common-sense statistic of taxes paid relative to income, or the average tax rate.

In fact, if the very richest people in America faced a marginal tax rate on their millionth-plus dollar earned of 70 to 80 percent, their average tax rate would still very likely be close to 30 percent. We might contemplate such high marginal rates at the top if we had failed to achieve the best solution of broadening the tax base and we were trying to make the tax system more progressive (while raising revenue for deficit reduction) by only raising — or creating new — top marginal tax rate brackets.

[But] Let’s be clear that I spent the whole hearing advocating for base broadening that would keep rates low. But I was asked what the maximum top marginal tax rate could be that the economy could handle, regardless of how successful or not we might be with base broadening efforts…

I encourage EconomistMom readers to view the hearing video and judge for yourself whether Senator Johnson was playing nicely or not.  Either way, I don’t think his or Dan Mitchell’s view that marginal tax rates on the rich are already high enough to be worrisome has much basis in reality.  (Nor did the story that came out the very day of the hearing (March 1) about Dan Mitchell’s organization, the Cato Institute, and how much it is influenced by the Koch brothers, help Mitchell’s credibility as an objective and fact-based economist.)

Even so, I still would prefer we raise revenue by not raising marginal tax rates further and instead broadening the tax base (reducing tax expenditures) in (very easily) progressive ways.  The “trilemma” of tax reform is entirely possible to achieve and is actually the best way to succeed, politically and economically, in doing good tax reform.

15 Responses to “The Tax Policy and the Economy Fairy Tale”

  1. comment number 1 by: AMTbuff

    I agree that the Laffer peak is somewhere north of 50% for wage income, although I doubt it’s as high as Diane’s 70% to 80%. For highly elastic income like realized taxable capital gains, the Laffer peak is probably below 40%. Considering that state income taxes and various phase-outs can add 10% to 17% to the official federal marginal rate, current federal tax rates are not far from revenue-maximizing levels.

    Furthermore if you look at the studies of revenue-maximizing rates you find that the Laffer peak is quite flat. That is to say, the government gets very little benefit from the last 10% increase in tax rate. (Similarly, the revenue lost if the government goes 10% past the Laffer peak is small.)

    Because higher than necessary tax rates cause significant harm to the economy, societal welfare is maximized at tax rates lower than the Laffer peak rates. Money is also useful to people when it’s in private hands, not just when it flows to the government.

    On the main point of Diane’s post, tax reform will be very beneficial once the public has decided whether it prefers to withdraw promised benefits from the non-poor or to dramatically increase taxes on everyone. Until the public votes one way or the other on this central question, adding revenues only delays the necessary decision and exacerbates the eventual fiscal crisis.

    Bailing out a leaking boat normally makes sense, but it’s not a good idea when you are bailing instead of patching the holes in the hull! First things first.

    Our leaders need to demand that the public decide now: Cut promises or raise taxes. Either option is far better than continuing to pretend that no decision is necessary.

  2. comment number 2 by: SteveinCH

    I’d like to respond to point 2 above…that tax rates need to be raised to balance the budget. This is simply false. A Congress that grew the budget at the same rate that the budget grew during the two terms of the Clinton administration (in nominal terms) would balance the budget within a decade under current tax policy.

    The fact that the old saw about you have to cut spending and raise taxes still exists is one of my greatest frustrations on this entire topic. The fact that people who study the problem like Diane continue to repeat it is an order of magnitude more frustrating.

  3. comment number 3 by: SteveinCH

    Should have said effective tax rates above to make it clear that it matters not whether the increase in rates is due to actual rate changes or changes in the structure of deductions, exemptions, and credits.

  4. comment number 4 by: B Davis

    AMTbuff wrote:

    Furthermore if you look at the studies of revenue-maximizing rates you find that the Laffer peak is quite flat. That is to say, the government gets very little benefit from the last 10% increase in tax rate. (Similarly, the revenue lost if the government goes 10% past the Laffer peak is small.)

    Can you provide links to some of these studies, specifically the ones that show that the Laffer peak is quite flat? I’m interested in this topic and would like to see the actual numbers. Thanks.

  5. comment number 5 by: AMTbuff

    B Davis, try this paper:
    http://www.ecb.int/pub/pdf/scpwps/ecbwp1174.pdf

    Figure 6 toward the back shows the Laffer curve for capital income. It shows a revenue peak at a tax rate of 37% (an after-tax ratio of 0.63). Considering that California taxes capital gains at 10% which is not deductible against the AMT, the combined federal and state capital gains tax rate for 2013 will be at 30% not counting the numerous rate bumps due to phase-outs. The revenue at a 30% tax rate is within 2% of its maximum attainable (Laffer peak) value. If this analysis is correct further tax increases on capital gains will be very costly to the economy for the revenue they gain.

  6. comment number 6 by: B Davis

    I encourage EconomistMom readers to view the hearing video and judge for yourself whether Senator Johnson was playing nicely or not.

    I did not have the time to watch the entire hearing but I did catch Senator Johnson’s comments starting at about minute 97. In fact, Senator Johnson appears to have been so pleased with his remarks that he has posted this portion of the hearing on his website and on youtube. I agree that he was not playing nice. He likely just wanted to get the panel members to commit to a number regardless of any specifics. If they gave a low number, he could then give some calculation to suggest that taxes were already above that level. If they gave a high number, however, he could accuse them of gouging people of their hard-earned wages. I notice that he jumped on the top of the Laffer range that you mentioned to say “so you believe that an American making a dollar should pay 80 percent…”.

    In addition, that first chart that he had held up was highly misleading. It listed the 10-year spending for the years 1992-2001, 2002-2011 (actual) and 2013-2022 (projected) to have been 16, 28, and 47 trillion dollars, respectively. Of course, the chart should have included the title “THESE NUMBERS ARE NOT CORRECTED FOR INFLATION AND ARE THEREFORE TOTALLY USELESS”. He then goes into a discussion of receipts as a percentage of GDP. This is, in fact, probably the most useful way to measure receipts and outlays. As a percentage GDP, outlays in 1992-2001, 2002-2011, and 2013-2022 averaged 19.9, 21.2, and 22.5 percent of GDP, respectively. And Summary Table S-6 from the budget projects that, from 2013 to 2022, discretionary outlays will fall from 7.7% to 5.0%, mandatory outlays will rise slightly from 14.0% to 14.4%, and net interest will rise from 1.5% to 3.3% of GDP. Hence, the big increase in spending will come from interest as rates rise from their current low rates. One other item of interest is that receipts for the three 10-year periods are 18.9%, 16.7%, and 19.2%, respectively. Hence, a part of the debt problem is the drop in receipts over the last decade.

    Another odd, if not misleading, comment was made by Dr. Mitchell after Diane. He pointed out that for much of our early history, the tax rate was zero and that zero would be the rate in his fantasyland. This would indeed be a fantasyland in that I suppose that the revenue fairy would have to pay for the armed forced and any other government services Dr. Mitchell feels are necessary. Also, he fails to mention that some taxes, like tariffs were much higher in the past.

  7. comment number 7 by: SteveinCH

    Just curious BDavis, do you actually believe that discretionary spending will fall from 7.7% to 5.0% of GDP?

    A more useful thing to look at is the growth rate of spending. From FY 1993 to FY 2001 (the two terms of the Clinton administration), spending grew in nominal terms by about 2.6% per annum. Were spending to grow at that rate for the next 10 years and current tax policy were to be unchanged, the budget deficit would be less than $100 billion in FY 2022 nominal dollar, inclusive of the same increase in debt service you describe above.

    But, despite what is in the President’s budget (a 4.5% annual rate of spending growth), actual spending growth will be even higher until such time as interest rates spike above the forecast rates which will, at some point in the future, undoubtedly happen.

  8. comment number 8 by: AMTbuff

    He likely just wanted to get the panel members to commit to a number regardless of any specifics.

    Correct. Both sides play this game. Ask a hypothetical question: “Is X theoretically possible?” If you say yes, you will be misquoted as wanting to do X. It’s gotcha politics, and it’s despicable.

  9. comment number 9 by: Arne

    “It shows a revenue peak at a tax rate of 37%”

    Do you think this has much validity given that the peak is well outside the range of the data they used to fit their model in the first place?

  10. comment number 10 by: AMTbuff

    Arne, I misinterpreted Figure 6. It shows the peak at a capital income tax rate of 63%, not 37%. This is not very different from the peak for labor income. What is very different is that you can bring in 90% of peak revenue at half the peak capital income tax rate, where the economic distortion is only 1/4 of what it would be at the peak rate. Therefore reaching for that last 10% of revenue appears to be a very bad idea.

    37% is very close to actual capital income tax rates used to calibrate the model, as shown in Tables 1 and 2.

  11. comment number 11 by: B Davis

    B Davis, try this paper:
    http://www.ecb.int/pub/pdf/scpwps/ecbwp1174.pdf

    Figure 6 toward the back shows the Laffer curve for capital income. It shows a revenue peak at a tax rate of 37% (an after-tax ratio of 0.63). Considering that California taxes capital gains at 10% which is not deductible against the AMT, the combined federal and state capital gains tax rate for 2013 will be at 30% not counting the numerous rate bumps due to phase-outs. The revenue at a 30% tax rate is within 2% of its maximum attainable (Laffer peak) value. If this analysis is correct further tax increases on capital gains will be very costly to the economy for the revenue they gain.

    Thanks for the link. In looking around on the web, I found an interesting article by Bruce Bartlett. He mentions a number of papers about the revenue-maximizing tax rate. He mentions the one you reference (an one-year earlier version, I believe) saying the following:

    Also in 2009, economists Mathias Trabandt and Harald Uhlig examined revenue-optimizing tax rates for the United States and Europe. They found that the United States is well below the revenue-maximizing top rate of 63 percent, that taxes on labor could be increased by 30 percent before labor supply dropped enough to reduce revenues from further increases, and that taxes on capital could be increased by 6 percent.

    Bartlett concludes his article with what I thought to be a thoughtful summary as follows:

    Of course, it goes without saying that the optimal tax rate in terms of revenue is not necessarily the one that maximizes growth or is socially optimal. Personally, I would prefer not to have a top income tax rate exceeding 50 percent, because it is important psychologically and morally that people not be forced to give more than half of their income to the federal government. However, given the magnitude of our nation’s fiscal problem, a rate higher than that may be inevitable unless the United States adopts a VAT, carbon tax, or other broad-based tax to supplement existing revenue sources.

  12. comment number 12 by: B Davis

    SteveinCH wrote:

    Just curious BDavis, do you actually believe that discretionary spending will fall from 7.7% to 5.0% of GDP?

    At the very least, following through on such a cut will likely be difficult. It would not be the first time that such projections proved overly optimistic. I blogged about just such a case 4 years in a blog titled The Decline in Spending Projected by the Bush Budgets. However, I believe that these cuts are largely mandated by the Budget Control Act of 2011 (BCA). That of course doesn’t mean that they will happen. But the budget pretty much has to assume that they will since they are mandated by current law. Similarly, the budget assumes savings in Medicare from the Affordable Care Act (ACA). By the way, I just posted the long-run budget projections (through 2085) at this link.

    A more useful thing to look at is the growth rate of spending. From FY 1993 to FY 2001 (the two terms of the Clinton administration), spending grew in nominal terms by about 2.6% per annum. Were spending to grow at that rate for the next 10 years and current tax policy were to be unchanged, the budget deficit would be less than $100 billion in FY 2022 nominal dollar, inclusive of the same increase in debt service you describe above.

    Can you provide a source for that statement? In any case, the graph at this link, shows that spending dropped sharply as a percentage of GDP in Clinton’s two terms. I don’t think that it’s realistic to repeat such a drop with the Boomer retirement just beginning.

  13. comment number 13 by: SteveinCH

    The link is from the CBO historical tables and using a calculator to get the growth rate. Although I just did the same calculation off the OMB tables and got a different number. Interesting.

    As to your last point, I’d suggest that at the beginning of the Clinton administration, many people would have said the same thing.

    As to your point about the BCA, do you remember Gramm-Rudman or the various PAYGO provisions we have had?

  14. comment number 14 by: AMTbuff

    Personally, I would prefer not to have a top income tax rate exceeding 50 percent, because it is important psychologically and morally that people not be forced to give more than half of their income to the federal government.

    Anybody whose income falls within the phase-out range who has two or more children in college faces a marginal rate exceeding 50% under current law. That phase-out range is well below Obama’s definition of “rich”.

    Most taxpayers have no idea how high their actual marginal rates are. Bartlett might not either unless he has run the numbers himself. This topic is rarely discussed in detail by tax experts. When it is, it’s usually the even more extreme situation of rates that exceed 100% at low incomes due to loss of free medical care and direct payments from the government.

    If the tax and benefit system were presented to the public openly and completely, the current marginal tax (and benefit reduction) rates would shock most people. As some economists have written (not in these terms), the economy benefits when the system tricks people into believing that their marginal tax rates are lower than they really are.

  15. comment number 15 by: B Davis

    One of the more hostile exchanges at the hearing was when Sen. Ron Johnson, R-Wis., questioned what we thought the maximum marginal tax rate should be. Each time [Len] Burman [of Syracuse, the other witness invited by Chairman Conrad] and I tried to respond that it depends on the breadth of and distortions within the existing tax base, Johnson interrupted and insisted on our providing a specific number without any qualifications. It was obviously a setup, as [Dan] Mitchell [the Republican witness from the Cato Institute] described in a blog post.

    Regarding Dan Mitchell’s blog post, I did leave a comment there containing most of my comment 6 above. Also, I remembered having seen one of Dan Mitchell’s papers written when he worked for the Heritage Foundation that reminded me of Senator Ron Johnson’s questionable chart. I’ve posted a blog about the topic titled How to Mislead with Statistics. Following is the conclusion:

    The above charts are just three examples of how data can be presented in a misleading manner. The first two present numbers that have not been corrected for inflation or population. In addition, they look at relatively short time spans. The last chart seems to be an example of cherry-picking. The data from 1977 to 1980 which goes counter to the conclusion that the author is proposing is simply deleted. In addition, the varying length of the time spans seems suspicious.

    These types of charts prompted me long ago to start crunching the data myself and presenting it on my website. Of course, everyone doesn’t have the time or inclination to crunch all of the numbers themselves. I only have time to crunch those that especially interest me. However, there are a couple of things that everyone can do. First of all, you can search out opposing analyses which may point out flaws in the data or the way that it’s presented. Secondly, you can demand verifiable sources for any data with which you are presented and, when possible, verify it. I can’t count the number of times I’ve come across data that is cherry-picked or just plain wrong. The only way to determine this is to ask for the source and verify the data. If the data is unsourced or sourced in a way that cannot be easily verified (such as “IRS data”), it’s probably best to ignore it. It may be unsourced or badly sourced for a reason.