The courage of nonpartisan analysis — an interview with Len Burman

Photo: Urban Institute

We need nonpartisan institutions, dedicated to objective analysis. That’s what the Washington, D.C.-based Tax Policy Center does — they score and analyze tax proposals so we can figure out what they’ll mean for taxpayers. But it’s increasingly difficult for nonpartisan organizations to get heard.

Unbiased analysis is a passion of mine. It’s also a passion of the respected economist Len Burman, an institute fellow and former director at the Tax Policy Center, and a former senior Treasury Department official in the Clinton Administration. (OK, he’s also my cousin, but that doesn’t make him any less smart.) If you want to know how Washington works, Len has seen a lot of it from the inside. So I interviewed him.

You have been the head of the Tax Policy Center for two stints totaling more than a decade. It’s a nonpartisan organization. Can you describe what it does, and how it’s possible to be a nonpartisan organization in Washington?

We created the Tax Policy Center to try to educate the public about taxes. Our theory was that good analysis clearly conveyed without any spin or bias would lead to better policy. It’s important because we need a working tax system to finance the government. Also, we use the tax system is a way of providing public services. For example, the earned income tax credit is the biggest cash assistance program for working age families. The tax code includes hundreds of billions of dollars of subsidies for health insurance, homeownership, and retirement savings. Do these subsidies make the economy work better or more fairly fairer? Are they worth the cost in terms of larger deficits and increased tax complexity?

We certainly can be nonpartisan, and we are. We have high level staff who have served in both Democratic and Republican administrations (and many others who have never held political appointments). Our advisory board is similarly diverse. Our only rule is that analysis must be evidence-based and we try to avoid language that reasonable people on the left and right might perceived as partisan.

Partisans are sometimes unhappy with us, but that is to be expected. Generally, they like us when our analysis supports their proposals and are unhappy when it doesn’t. In the same TPC session in which a Trump representative, Peter Navarro, attacked us for being partisan and incompetent, he also read approvingly from the parts of our reports that he liked (including critiques of Hillary Clinton’s proposals).

What do you think are the stupidest elements of the tax code, that would be easiest to fix?

There are lots of stupid elements and none of them is easy to fix politically. The alternative minimum tax (AMT) is the poster child for pointless complexity. After you have calculated your income taxes under the regular rules, you must calculate taxes under a different set of rules and tax rates. If that calculation produces a higher tax bill, you pay the difference as AMT. It was originally intended to rein in tax shelters, but it mostly hits upper middle income households. If Congress thinks that certain provisions produce an unwarranted tax benefit (and some certainly do) it should just repeal or modify those provisions. A bunch of phaseouts in the tax code are complicated and confusing. For example, itemized deductions are reduced for taxpayers with high incomes. It is actually an obfuscated income surtax (about 1.2% for taxpayers in the top bracket), but confused taxpayers think it reduces the tax savings associated with charitable deductions. The phaseout might reduce donations, which is clearly counter-productive. It would be much better—and simpler—to ditch the phaseouts and explicitly adjust tax rates.

Can you describe in a general way what the likely effect of Trump’s tax plan would be on families and businesses in the US?

Trump’s tax plan would cut marginal tax rates on individuals and businesses, with most of the benefit accruing to those with very high incomes. It would cut federal tax revenues by more than $6 trillion over the next decade. It would move the tax system in the direction of a consumption tax, which many economists think could boost growth. It would cut tax rates on new investment, which would also stimulate the economy in the short run, but unless spending is cut dramatically, rising debt would push up interest rates which would eventually deter investment and harm the economy. The plan would also eliminate the estate tax.

When you did your analysis of Trump’s tax plan, his campaign attacked you. Is that unusual? What do you think it says about the changing political climate?

It was unprecedented. We have heard privately from campaigns in the past—both Democratic and Republican—that they were unhappy with aspects of our analysis, but I think most understood that we could not conform our analysis to campaign spin.

On one level, we take the attacks on TPC as a sign of our success. If people weren’t paying attention to our analysis, we wouldn’t be worth attacking. The Trump campaign’s attacks are on another level. I am hoping they are idiosyncratic to the Trump campaign and don’t say anything about the changing political climate, but only time will tell.

You were a senior official in Bill Clinton’s Treasury Department, under Treasury Secretary Larry Summers. What was it like to be a senior government official? Was was the best part of that? What was the worst?

One of my favorite songs in the musical, Hamilton, is “Room where it happened,” expressing a yearning to be in the room when important policy decisions are made. I got to be in the room where it happened for every tax proposal in the last two years of the Clinton Administration (and, surprisingly, there were a lot). I got to shoot down a lot of bad ideas and, even in cases where my wise counsel was rejected, I had my say. I had a fantastic staff and I learned a huge amount about the tax system. The experience gave me the credibility and knowledge to be able to start the Tax Policy Center.

The worst part: being on call 24/7, which took a toll on my family life. Next worst: when I lost the internal policy argument, I sometimes had to give speeches supporting policies that I thought were ill advised. I never said anything that I thought was false, but spin is a form of deception. I didn’t like that.

Donald Trump has come up with some very dumb and dangerous ideas. But let’s stick to tax policy. Which of his ideas do you think would be the most dangerous for the economy, and why?

 $7 trillion in additional debt at a time when the retirement of baby boomers and rising healthcare costs will put unprecedented demands on the federal government is a really bad idea.

Why can’t economists talk in language that regular people can understand?

I like to think that some of us can. Part of the problem is that we use mathematics and models to make sense out of complex economic systems. Sometimes that stuff is just hard to explain. We have a language that is very efficient for communicating within the profession. Translating for people that don’t have the background in economics either requires a lot of explanation or a lot of simplification. Our peers will often take us to task if we oversimplify.. “You said X, but X isn’t true under certain circumstances.” (We also like to use symbols, like X.)

That said, I think it’s important to communicate with real people, particularly if you want them to support better policies. I am willing to oversimplify a bit, especially where simple stories are consistent with empirical evidence, but I also think it’s important to cite the rigorous research that supports the anecdotes. That’s why I love hyperlinks. I can present a simple intuitive story with links for those who want to examine the rigorous underpinnings of my analysis.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.


  1. “The tax code includes hundreds of billions of dollars of subsidies for health insurance, homeownership, and retirement savings.”

    True, but the Treasury by law recovers essentially 100% of the billions of dollars in subsidies that go to retirement savings. The recovery happens over time through voluntary taxable withdrawals and, after age 70 1/2, mandatory minimum required distributions. Voluntary or mandatory, withdrawals are taxed at ordinary income rates (not the lower capital gains rate). Over the years, account holders and their heirs totally repay the subsidies they’ve received via pre-tax contributions and tax-deferred capital gains. Putting it another way, in the end the retirement tax break effectively costs the Treasury nothing. You could easily go even farther and call the retirement tax break a sure-fire investment.

    1. Gerald, The subsidy for retirement savings isn’t the up-front deduction, which as you note is repaid with interest. The subsidy arises from the fact that the earnings on retirement accounts are untaxed. If you took the same amount of after-tax earnings and put it in a taxable account, and interest, dividends, or capital gains would be taxed under an income tax. Exempting capital income from tax is a big subsidy (compared with a comprehensive income tax).

      Basically, retirement savings are taxed as under a consumption tax. Some believe that should be the baseline for comparison. But if the baseline is an income tax, the tax expenditure is significant.

  2. “The subsidy for retirement savings isn’t the up-front deduction, which as you note is repaid with interest.”

    Besides the up-front deduction, i.e., the pre-tax contributions, account holders also pay the Treasury for those tax-deferred capital gains; taxes are ultimately paid on the total holdings in the account. What am I not seeing here?

    “The subsidy arises from the fact that the earnings on retirement accounts are untaxed.”
    In this context, what’s the meaning of “earnings”? I’m a (very fortunate) retirement account holder, and the only “earnings” I’m aware of are, 1. Capital gains prior to withdrawal, on which I pay taxes via my withdrawals, 2. Capital gains which might accrue after withdrawal, when I might reinvest the money. On gains in non-retirement accounts, of course, I pay the regular capital gains taxes. Leaving the question: what “earnings” are “untaxed”?

    Thanks for bearing with me.

  3. Simple example. You earn $1,000 and there is a flat 25% income tax rate. Suppose you can invest it in an account that pays 4% per year in interest. You withdraw the entire amount after 20 years (when you are over 59 1/2 years old) and pay tax at 25%. The after-tax value of the account is $1,643. (The general formula is X*(1+r)^N*(1-t), where X is the pre-tax contribution amount ($1,000 in this case), r is the interest rate (4%), N is the holding period (20), and t is the tax rate (25%).) Note that if you had put the after-tax earnings, $750, in a Roth IRA, you’d have the same retirement income, which is why Roth IRA and Traditional IRAs are equivalent when you contribute the same amount of after-tax income and the tax rate is the same during contribution and retirement.

    Now, suppose you put the same amount of after-tax income in a taxable account. You have $750 to invest after paying the $250 in tax. You earn 3% interest after tax (4% less 25% tax). After 20 years, you will have $1,355 in the account. The difference between the $1,643 you’d have in the IRA and the $1,355 you’d have in the taxable account is the tax subsidy. (The formula in this case is: X*(1-t)*(1+r*(1-t))^N.)

    If you’re still not convinced, look at this nice explainer by my Tax Policy Center colleague, Lydia Austin:

    1. From Lydia Austin’s explainer: “Most people think of IRAs as tax-deferred. After all, you must pay tax when you withdraw money from the accounts. But to an economist, IRAs are tax-free.”

      I understand Lydia’s explainer, but I’m with most people: IRAs are tax-deferred, not tax-free. The example compared IRA retirement accounts (pre-tax contributions) to Roth IRAs (post-tax contributions), and included a chart showing that both accounts, after 30 years, would have the same totals.

      That’s a mathematical calculation, but here are two non-mathematical counter-arguments: 1) Roth critics* have correctly maintained that, over the long term, Roths are a big loser for the Treasury; Roths in fact will *never* repay the Treasury for any of the forgiven capital gains taxes; 2) Whatever the “subsidy” to retirement accounts, they are (in fact and in intent) an investment by America in the retirement security of seniors. They are also, via both voluntary and mandatory withdrawals, a significant economic stimulus: over $234 billion in 2014, per the latest IRS figures. Does Lydia’s explainer take into account all the taxes (federal, state, local) that will utimately be paid as a result of that $234 billion in the pockets of retirees?) *Including Len Burman.

      1. You don’t understand Lydia’s explainer or mine. Roth’s are equivalent to traditional IRAs from the point of the individual (again, assuming equal contribution of pre-tax dollars and a tax rate that doesn’t change over time). If you buy that Roth accounts are tax free (which seems obvious to me because there’s never any tax), then the equivalency means that Traditional accounts are tax free too.

        This is not a matter of opinion or interpretation. The equivalence is mathematical. (I suppose one could posit an alternative mathematical system, but I’m pretty sure you’re not buying into alternative facts.)

        There is are meaningful differences for the Treasury. (1) Effective contribution limits to Roths are higher than Traditional IRAs, which means that more income can be sheltered; (2) As you note, the timing of receipts is front-loaded, meaning that switching from Traditional to Roth accounts exacerbates our long-run budget problems; and (3) Roth IRAs earn a higher risk-adjusted rate of return than Treasuries, then the Treasury loses by not being able to share in those gains (as it does with a Traditional IRA). Also, Roths do not require withdrawals and offer estate tax benefits because pre-paying the tax reduces the size of one’s taxable estate. That makes Roths more attractive to people who do not actually need retirement accounts to finance retirement consumption.

        As to whether the subsidy is worth the cost, that depends on how much it actually increases saving. The tax benefits might encourage people to save more or less (income and substitution effects). For very high-income people, there’s is no marginal incentive to save more because they save much more than the contribution limits. For them, the tax windfall probably increases current consumption (income effect). The only thing that has clearly been shown to increase retirement saving is employers choosing to deposit a part of wages into the accounts on employees’ behalf as the default option.

        1. Thanks for the dialogue; I do appreciate it.

          “There are meaningful differences for the Treasury.” Yes, there are. In addition to those you mentioned, there are the tax inflows I mentioned earlier (Treasury, state and local) resulting from the voluntary and mandatory withdrawals from non-Roth retirement accounts. Because Roths are not subject to mandatory withdrawals, the inflows from Roths are necessarily far smaller. (Correction from my earlier comment: ultimately, Roth account holders or their heirs will ultimately pay taxes on their total account holdings.)

          “As to whether the subsidy is worth the cost, that depends on how much it actually increases saving….For very high-income people, there’s is no marginal incentive to save more…” Understood; but for people with less than high incomes (myself, way back when) there was definitely an incentive to save more, and this was before 401(k)s and the employer contribution. I can’t believe that I was an outlier in responding to that incentive.

          What if I reworded the beginning of the previous graph and put it this way: “As to whether the subsidy is worth the cost, that depends on how much it helps to increase the retirement security of America’s labor force.” I think that was the intent of Congress in passing ERISA back in 1974, and the bill succeeded in achieving its purpose; there’s more to do (over half the labor force still lacks workplace access to retirement accounts), but the foundation Congress laid was a sound one (and not, all told, a costly one).

          1. Memo to myself: Before sounding off publicly, please do the public a favor: don’t be in a rush and write stupid things. Because I didn’t follow this rule, the comment above says that “Roth account holders or their heirs will ultimately pay taxes on their total account holdings.” *False*; no taxes are paid on Roth withdrawals. What I meant to say is that Roth accounts, while they lack the minimum distribution requirements of regular retirement accounts, must ultimately be distributed. That’s a far different proposition than paying taxes. P.S. To Len Burman, thanks again for the enlightenment and for Lydia Austin’s explainer; I think it’s finally getting into my thick skull.

  4. This is an important topic as escalating deficits and debt are poorly understood by both the public and, sadly, economically ignorant members of Congress. My favorite source for bipartisan research on this subject is:
    Committee for a Responsible Federal Budget
    1900 M Street, NW
    Suite 850
    Washington DC 20036 United States