Tuesday, December 19, 2017

How to cut taxes and raise tax rates

How can you cut taxes but raise (distorting, marginal) rates at the same time? Add a deduction, but phase it out with income. Then people below the income limits pay less taxes. But as the income limit  phases in, the marginal tax rate is higher than the previous rate. The new (and old) tax code is full of this perverse result.

For example, suppose you start with a tax code where everyone pays 50% of income. Then, add a deduction, credit, or exemption so people who earn, say, less than $100,000 of income pay no taxes. But phase it out over the next $100,000. Thus, people who earn $200,000 pay the original 50%, or they pay $100,000 of taxes. People who earn $100,000 pay no taxes. So, we have engineered a 100% marginal tax rate for people between $100,000 and $200,000 of income -- each dollar is completely taxed away!

In my example, we gain a 0% (down from 50%) marginal tax rate for people below $100,000 of income. But if the $100,000 is a fixed deduction or credit that does not scale with income, even that benefit is lost.

"Tax cuts" are not necessarily good for growth! It is possible to cut taxes and raise marginal rates, reducing growth.

This came to mind while reading the interesting "Games They Will Play"
Individuals who provide “specified services” (such as lawyers and doctors) must have taxable income of less than $315,000 for a married couple (or half that for a single individual) to be fully eligible—with the benefit phasing down over the next $100,000. 
"Games they will play" makes no mention of this or any other marginal rate. As is common in tax analysts they are great on disincentive margins to game tax payments by reclassifying income, but not so good on these marginal incentives.

I would love to see a true marginal analysis of the tax proposal. What are its actual incentives and disincentives, when you put it all together,  not the constant who-gets-what commentary.

"Games They Will Play" is good reading if you have half a mind to pick up your pitchfork and join the other peasants in rebellion. It's phrased as problems with the new tax code, but it gives you a great condensed sense of just how rotten the old tax code is.

These are essentially unavoidable complications resulting from an income tax.  Once you have an income tax you must have a corporate tax at roughly the highest individual tax rate, or people incorporate. "Games they will play" focuses on this margin with respect to pass-through businesses, and the hopeless quest to separate investment in pass throughs from what is essentially wages. Once you have an income tax you must have a capital income tax, or people turn wage income to capital income -- take stock options rather than wages. Once you have capital income taxes you have to tax capital gains, or companies don't pay dividends and instead pay capital gains.

"Income" is a very poorly defined concept, once you get past the basics of people who earn wages and consider professionals, top management and small business owners.

And so on and so on. This is what leads me to a uniform VAT instead of an income tax. It really is the only way to get rid of the mess. That really is the lesson of the current tax bill. 31 years of waiting (since 1986), thinking, writing papers, opeds, blog posts, think tank reports, articles, and we get... this. It must be inevitable given a few basic starting points, and a progressive income tax is that starting point.

When you object, consider, is this apparently unavoidable mess,  insane complexity, and obvious politicization worth whatever benefit you see from taxing income rather than consumption?

I agree with the WSJ editorial page, in the context of the income tax, the only sensible way to reduce the pass through corporate rate is together with a much lower -- it mentions 28% -- top individual Federal tax bracket. Then soak the rich by getting rid entirely of mortgage interest, charitable, employer health care deductions, electric car credits, and so forth.   

There is much discussion on the left that once you allow companies to completely expense investment, you can have a large corporate tax without disincentives to invest. I'm not convinced, as this sets an army of tax lawyers out to define what is an expensable investment and what is not. Obviously, if companies can deduct "investment" in stocks and bonds, then there is nothing left to tax. The tax lawyers who write the code are smart enough to stop that. But really, how different is it to "invest" in a forklift than to buy stock in another company? (And then rent the forklift out to someone else.) To "invest" in a fleet of company Ferraris, jets, apartments, and so on? On the other side, much corporate investment these days is in "intangible" capital, including software. If I hire people to build out a new IT system, or a better inventory management system, this is economic investment. I don't see how to expense those, and leave anything behind to tax. 

So what is the idea? Somehow, the gnomes will separate "rents,",pure profits, from normal returns to capital and still tax the "rents" leaving untouched the incentives to invest in actual measurable and unmeasurable capital. 

That seems pretty hopeless.  A low (zero) corporate tax seems much simpler. Tax people when they spend the money. 


  1. I agree with this post.

    Though I do think a simpler national sales might work rather than a VAT, when added to a national property tax, a national pollution tax, national fossil fuels tax, some national Pigou taxes (yes, I know, my virtue is your Pigou), and across-the-board import tariffs.

    If these taxes, none of which tax income or productive behavior. are not enough then add on a simple flat 10% on on reported incomes above some high threshold, such as $1 million.

    Yes, the flat tax on incomes above $1 million is class warfare, but then the Washington establishment is class warfare, and guess who usually wins and who wrote the tax code we now have?

    Radically cutting back the federal government, starting with the $1 trillion a year in "national security" outlays might be necessary too, although I am not entirely convinced on the demerits of helicopter drops.

  2. John,

    "And so on and so on. This is what leads me to a uniform VAT instead of an income tax. It really is the only way to get rid of the mess."

    And again I will say the mess isn't created from having an income tax. It is created because we have 435 House Members and 100 Senators all with their fingers in the pot, all trying to please constituent groups, and all trying to get re-elected.

    The only way to fix this mess is to take Congress out of the loop in terms of making tax policy changes.

    The model to look at is the Federal Reserve Act of 1913.


    Prior to 1913, U. S. monetary policy was set by directly by the U. S. Congress.

    "When you object, consider, is this apparently unavoidable mess, insane complexity, and obvious politicization worth whatever benefit you see from taxing income rather than consumption?"

    That is a bogus question. You have offered nothing that demonstrates that a VAT or national sales tax will over time remain free of the same complexities that permeate the U. S. income tax code.

    1. Though you get a C- for politeness, this is an interesting political observation. Perhaps the best route to serious tax reform in a democracy is to put the details in some independent structure. There are ways short of an independent agency. Quite often Congress appoints a bipartisan commission to come up with a plan, and then commits itself to an up or down vote. Base closing and trade treaties are done this way. Several tax reform proposals were almost done this way, most recently the Bowles-Simpson effort. A full permanent independent agency is more likely to be captured, as a business wanting tax breaks can bring a lot of pressure to bear on a commission bureaucrat. Financial regulation is hardly a poster child for simplicity, transparency or effectiveness, yet it has been run by independent agencies including the Fed for a long time. Still, you have raised an important question -- the two founding assumptions are an income tax, and that the detailed nature of taxes will be written by congress itself, rather than by a commission or a regulatory agency. I think we can agree that getting a serious reform ought to consider both the economic and the political assumptions.

    2. John,

      "Though you get a C- for politeness..."

      You got the nickname "Grumpy Economist" for polite conversation?

      "Quite often Congress appoints a bipartisan commission to come up with a plan, and then commits itself to an up or down vote."

      And many times those commission reports end up on a shelf collecting dust - Simpson Bowles being the latest example.

      "A full permanent independent agency is more likely to be captured, as a business wanting tax breaks can bring a lot of pressure to bear on a commission bureaucrat."

      And yet the U. S. central bank (Fed) has been able to avoid "capture" because it doesn't deal directly with businesses that are looking for cheap financing. Instead the central bank makes loans to private banks who in turn can approve / disapprove loans to said businesses.

      Independent tax policy (by commission or agency) must also include intermediaries that deal directly with businesses / individuals (in the same way that private banks act as intermediaries between the central bank and borrowers).

      "Financial regulation is hardly a poster child for simplicity, transparency or effectiveness, yet it has been run by independent agencies including the Fed for a long time."

      Compare the Federal Reserve Act to the U. S. tax code and tell me which is a simpler regulatory structure?


      The Federal Reserve Act (as amended to date) consists of 31 sections.


      The federal tax code consists of about 6800 sections divided across 11 subtitles. The income tax code alone consists of over 1500 sections.

      "I think we can agree that getting a serious reform ought to consider both the economic and the political assumptions."


  3. Thanks for the this. It's a compelling vision.

    I've been reflecting over the past few days, and my greatest remaining barriers to a tax system that is only taxes consumption are:
    - Distributive impact - how would you make taxation progressive (assuming you think that's appropriate)? Would you do VAT + UBI? Some form of progressive consumption tax?
    - 'Excluded beneficiaries' - do you think there are people who benefit from the government expenditures who wouldn't be taxed? e.g. foreign owners of US companies
    - Transition - how would you start down this road? Potentially this was the problem with DBCFT. Transitioning all at once seems very difficult (it would hurt some sensitive political constituencies, FX impact could have a wide blast radius), but can you have an initial stage of a low VAT while being worth the administrative costs?
    - Collection efficiency - Taxing a single type of transaction adds incentives to avoid / evade it. OTOH VAT is fairly difficult to evade.
    - Incentive to capture - Do you have any concern that large pools of inter-generational wealth will be a) faster to grow and b) incentivized to capture policy in order roll-back a consumption tax?

    Do you have any views / counter-arguments on these?

  4. Great post, as always. Another option (as an alternative to a VAT) would be a comprehensive wealth tax, which David Shakow and Reed Shuldiner from Penn proposed a while back. I haven't read it in a long time, but I believe this is the paper: http://scholarship.law.upenn.edu/faculty_scholarship/1264/

    A VAT is still probably a bit easier, whether through a credit-invoice model or an unlimited savings account (USA) model; but this is an intriguing alternative.

  5. I agree on the merits of a progressive consumption tax. But what is the best way to implement this?

    You propose a VAT with (I assume) a universal basic income.

    Let’s contrast this with a progressive personal expenditure tax, i.e. a tax on wages plus net withdrawals from wealth.

    What are the main compliance issues in each case?

    1) Under both systems, the government needs to monitor businesses to ensure that they do not classify private consumption of American citizens as if they were legitimate business expenses. As you say: “a fleet of company Ferraris, jets, apartments, and so on”. This is an unavoidable feature of pretty much any tax system.

    2) A VAT requires border adjustments, i.e. cash inflows to a company from foreign sources (exports) are untaxed, while cash outflows to foreign sources (imports) are not deductible. This creates incentives to misclassify the source/destination of a cash flow, and will require careful monitoring.

    It seems to me, then, that a PET requires less government effort to mitigate evasion than a VAT.

    Note that neither system requires the next-to-impossible task of disentangling a company’s labor from capital income, as you say: “the hopeless quest to separate investment in pass-throughs from what is essentially wages”. For instance, if you incorporate under a PET and sell your labor services, you will be able to deduct the cost of investments in your mini company. But income subsequently earned (and consumed) is simply taxed at ordinary rates. (This is equivalent to a system with complete investment expensing: either you separately subsidize (tax) each corporate cash outflow (inflow); or you subsidize capital raised, but tax dividends and buybacks.)

    Finally, I think slowly moving in the direction of a PET is far more politically feasible than switching to a VAT. For one thing, a VAT would likely be a pretty stiff tax on existing capital (the elderly will not be pleased). Plus, a UBI has so far not proven to be a popular idea.

    1. See https://www.wsj.com/articles/tax-consumption-through-a-vat-and-voila-1504550331
      on how to implement a progressive VAT. But we should examine the progressivity of the entire government, not the tax system in isolation. We already have UBI, in the form of refundable credits. Whether we implement a consumption tax via the current code -- measure income, give credit for savings -- or directly -- measure consumption, don't bother trying to define and measure "income" and "savings" is really a second order issue. Compared to the current mess it's which cloud of free market nirvana do you want to live on.

    2. Gideon,

      The reasons that a progressive Personal Expenditure Tax won't work are many fold.

      Reason #1: Personal expenditures can be financed by borrowing. The very basics of tax policy would tell you that a person should only be taxed based upon what they can afford to pay.

      "Let’s contrast this with a progressive personal expenditure tax, i.e. a tax on wages plus net withdrawals from wealth."

      Net wealth can go negative through borrowing. Sure you could try to pay taxes owed on borrowed purchases through additional borrowing, but a bank can turn down the second loan application.

      Reason #2: A progressive tax on consumption requires the government to have some forehand knowledge of the total consumption that a person will partake in over a year.

      John's solution is end of tax year refunds which have their own problems - namely that they create an incentive to postpone consumption until close to the end of the tax year. Why would I pay full tax in May of 2018 to have a portion of that tax refunded in April of 2019 when I can simply defer all of my 2018 consumption purchases until February of 2019.

      Income taxes (and specifically withholding on income taxes) do not suffer this problem.

    3. I'm not sure why #1 is a problem. If I take out a loan I will have to service it somehow. If I use my labor income to do so, it will be taxed, if I withdraw from my wealth ditto.

      I don't understand #2: why would the government be much better at predicting a person's annual income than their annual consumption? And why does it matter if it is imperfect? Just make up the difference at tax time. We could abolish withholding and the system would still work fine.

  6. I think your statements about the intellectual difficulty of defining and taxing income are correct. However, I think using a VAT as cure all is politically hopeless. Most European countries adopted the VAT because their income taxes would not generate enough revenue to pay for their social welfare programs. But, they kept their income taxes too. The US did not adopt a VAT because it did not need to.

    Even though the US welfare state is growing like Topsy and the Republicans have ratified the existence of Obamacare, the parties are dead set against a VAT. Republicans hate it because they still hope to starve the beast. Democrats refuse to consider it because it will mean that almost everybody has to pay some taxes, in violation of their mantra: "don't tax you, don't tax me, tax the man behind the tree".

    If we did adopt a VAT, and we wanted to preserve the sort of spiteful progressive taxation we have, we could re mold the income tax as a tax on consumption, by allowing unlimited IRA type accounts. The biggest problem with that is the treatment of borrowing. In a true consumption tax, net borrowing should be taxable. But, the automobile and housing industries would be cataleptic if you proposed that. Maybe you could work out a an amortization scheme. E.g. take ten percent of the cars price every year as income.

    Personally, the process of adopting this latest set of changes to the tax law has been very instructive. I have given up any idea of a true intellectual simplification of the tax code. Just about every provision has a an enormous constituency that will fight tooth and nail to preserve it. I was quite impressed that they could cut back on the SaLT deduction and the mortgage interest deduction to the extent that they did. If they really managed to kill the HELOC deduction, they did God's work.

    I think the changes they made to the corporation tax were good and timely, the US needs to be competitive. We are not helped by having US businesses incorporate in foreign countries. The devil, of course, is in the details, but I am inclined give it a chance to work out before reassessing.

    1. I have been looking for information on the contents of the act. Apparently the HELOC deduction is dead. Praise the Lord.

  7. Another thought from the process. I have been listening to politicians complain about tax cuts for the rich for as long as I can remember. You would think that capping the SaLT deduction at $10,000 would make them happy, but no.

    It is also remarkable to me that the the politicians who screech the loudest about inadequate taxation of the rich represent the most affluent places in the country. If they are truly representative of their constituencies, we should bow to their wishes and give the people what they want. I think a top marginal rate of 50% on the famous 1% ($355,200 single $502,300 couple) would be appropriate. Just to show how nice I am I would abolish the AMT and the Obamacare investment income tax.

    We might then discover the true impact of marginal rate increases, and, more importantly, whether politicians really believe what they say.

  8. It seems countries compete, more or less, for (K) capital. Lowering the corporate rate to 26.5 would attract more foreign K. The fly in the ointment, Debt to GDP is expected to increase. Will that be enough to forestall default? The issue of fiscal discipline, lack thereof, and myopia are concerns


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.