Congressman Donald Schwerbitz, who represented South Dakota back in the 1960s and 70s, was a visionary environmentalist who sponsored the first legislation designed to reduce our national carbon footprint. It was Congressman Schwerbitz who recognized that carbon emissions are caused primarily by breathing, and he proposed to cut those emissions in half by requiring every American to wear a device that plugs up one nostril.
Congressman Schwerbitz was actually my neighbor Les Waas, the founder of the Procrastinator’s Club of America (which seems not to have gotten around to launching a website) and an irrepressible prankster who managed to get himself invited onto a talk radio program to explain how the nostril plugs would work. (The host was in on the joke.) Because talk radio audiences are dominated by libertarians and reactionaries, the response was not positive. Callers clamored for civil disobedience; one threatened that if he ever saw anyone wearing one of these devices, he’d “punch him in his other nose”. Others worried that our clean air might drift over to Cuba, where the communists could use it. A few, though, were enthusiastic. One woman wanted to know if the devices could be adapted to fit animals. Warthogs, she observed, have very big nostrils.
All of which makes me a little hesitant to comment on Len Burman’s op-ed piece in yesterday’s Washington Post. It’s hard to read the column without suspecting that Mr. Burman is a latter-day Senator Schwerbitz looking to entice bloggers like me into making fools of themselves by taking him seriously. But okay. I’ll take the risk.
Mr. Burman likes President Obama’s proposal to freeze (some) federal spending. He likes it so much that he thinks we should apply it more generally. In order to do this, he wants to count not-taxing-stuff as a form of spending. For example, we currently don’t tax income that’s used to pay mortgage interest or employer-sponsored health insurance. (I happen to favor the first of these policies and oppose the second, but for the moment that’s neither here nor there.) Mr. Burman refers to these failures-to-tax as “tax expenditures” and thinks we should freeze them as well.
Of course there are a lot of other things we don’t tax, and the “logic” of Mr. Burman’s op-ed requires us to count all those lost opportunities as expenditures also. The Federal Government currently spends $300 billion a year on not levying a $1000 birthday tax. It spends tens of billions more on not taxing breathing. And only the Congressional Budget Office can estimate the cost to taxpayers of not taxing taxpaying.
Look. It’s really important to distinguish between spending on the one hand and not-taxing on the other. Spending consumes resources, which are then unavailable to members of the next generation. Not-taxing doesn’t do that (though you could argue that not-taxing sometimes encourages individuals to consume resources, which are then equally unavailable to members of the next generation).
Crazy as all this may sound, it’s not out of the mainstream. Deficit hawks often conflate tax cuts with spending increases, as if one were equivalent to the other. That’s silly, and I’m grateful to Mr. Burman for making the silliness graphic by taking it to such an extreme. I hope that was his intention. But I fear he was serious.
“Tax expenditure” concept can actually make sense in some cases. For example, what is the difference between the following policies:
1. All farm income below 100 thousand is not taxed at 20% rate (lets assume the flat rate)
2. All farmers get 20% of their income (up to 100 thousand) as “income support subsidy”
In the real sense there is no difference in these two policies, but in budgetary terms there is. The first one reduces both the revenue and expenditure side of the budget, the other one increases it. If the government wants to “cut spending”, it could provide tax cuts to their friends and constituencies instead of providing subsidies. The requirement to make those things public as “tax expenditure” increases the transparency and could be used as a deterrent to engage in such activities and makes it also makes trade-offs more visible (do we want to provide more tax incentives to investors, or to tax them and invest that money in infrastructure, for example).
The point is that “tax expenditure” concept probably makes sense when you want to take account of all the unusual deviations from a normal tax regime. I agree that “unusual” is not a technical or scientific term, but I guess you know what I mean. “Not taxing breathing” is not one of those, but “10 year tax holiday for new investors” probably is.
Believe it or not, IMF is insisting that my country amends the Budget System Law in a way that would provide information on “tax expenditures” and it seemed to me to make some sense.
I hesitate to say this, but, after a quick look, Burman’s article kind of makes sense to me. Perhaps I misunderstand it, but what Burman seems to say is that we should eliminate tax breaks: tax everything at the same rate, as you (Steve) have been insisting for the last couple of weeks; no more micro-management of the economy through the tax system.
BTW what is wrong with abolishing mortgage tax relief? isn’t m.t.r. a factor behind housing bubbles? (not to mention the feudalization of the economy.) If you have a mortgage, I think that you should declare an interest before defending mortgage tax relief.
Also, while it would be good to tax employer-sponsored health insurance, it might make sense to provide a tax exemption for health insurance bought by individuals or families.
Snorri:
BTW what is wrong with abolishing mortgage tax relief?
I’ve argued elsewhere (and I think you’ve agreed) that interest income should not be taxed. On mortgages, the income earned by the lender *is* taxed. The burden of this tax is partly passed on to borrowers through higher interest rates, so it’s split between the lender and the borrower.
By giving a mortgage exemption to the borrower, one basically undoes the effect of taxing the lender. The benefit of the tax break is partly passed on to lenders through higher interest rates, and is therefore split between the lender and the borrower. With a little more work, you can show that this benefit is split in the same proportion as the tax burden, so everyone gets back exactly what the mortgage interest tax is costing them.
So the mortgage interest deduction is equivalent to not-taxing the mortgage interest earned by lenders, which is the right policy.
There is an extensive literature on this question in the law reviews. While it is indeed absurd to treat all failures to tax as expenditures, some failures to tax are functionally identical to expenditures and should be viewed as such. To be sure, there are cases in which the classification of a failure to tax as simply part of the tax system or as a “tax expenditure” will be controversial. The mortgage interest deduction is one of them: I once wrote an article arguing that it shouldn’t be viewed as a tax expenditure; most tax lawyers would disagree.
Steve: do you mean “The burden of this tax is partly passed on to borrowers through HIGHER interest rates”?
Just trying to clarify – if you did mean “lower”, can you please explain how?
Thanks!
Dave
Dave: I meant higher and I am editing to fix this.
I very much like this seemingly silly but surprisingly respected concept of “tax expenditure”. Burman is wrong in his estimate of how many there are: our tax expenditures are infinite, as follows from Steven’s remarks. But our infinitude of tax expenditures is infinitely wise and productive. For example, our tax expenditure on the breathing of our citizens: that’s a wise expenditure, for it encourages our citizens to breathe, and no one can argue with the desirability of that. Similarly for the birthday tax expenditure, for it encourages our citizens to continue having birthdays. If only Burman could find a way for us to tax the breathing and the birthdays of the non-citizens, then he would be on to something real.
The mortgage interest deduction is equivalent to not taxing the current consumption of housing services of homeowners. It is a tax distortion of the type you railed about earlier.
Tax expenditure. Suppose the State of California wants to attract more teachers. One, it could pay them more–more budget expenditure. Two, it could exempt teacher’s salaries from taxation–more “tax expenditure”. Both achieve the same end. Both require that other taxes be increased or other expenditures be cut. How are they different?
Neil: The mortgage interest deduction is equivalent to not taxing the current consumption of housing services of homeowners. No, it’s not, because only homeowners with mortgages get the deduction.
Suppose the State of California wants to attract more teachers. One, it could pay them more–more budget expenditure. Two, it could exempt teacher’s salaries from taxation–more “tax expenditure”. This one is a real expenditure, but it’s one any economist would have counted in the first place.
@Steve_Landsburg: You make a good point here, but you marr it by unnecessarily revealing your confusion about CO2 and breathing. Had you become informed on that particular issue, you would know that breathing is carbon neutral, because all the CO2 you exhale had to come from plants, which got it from the atmosphere, so you’re only returning to it, that which was already there.
(Yes, the same happens from fossil fuels, except that their trapped CO2 from eons is being concentrated over a very short interval.)
Yes, environmentalists propose stupid policies, but restrictions on breathing do not fall out as an implication.
Silas: Still no improvement in those reading comprehension skills, I see.
Neil: The mortgage interest deduction is equivalent to not taxing the current consumption of housing services of homeowners. No, it’s not, because only homeowners with mortgages get the deduction.
The homeowners without mortages already enjoy their “imputed rent” tax free.
What I’ve never quite understood is why (state school) teacher’s salaries are taxed in the first place. They are given money by the government, and then the government takes some of this money back. It would seem simpler not to give it to them in the first place. Presumably the reason is to allow easy comparison between salaries in the private and public sectors, but it still seems little more than an accounting trick.
Looked at from this direction, it seems fairly obvious that exempting teacher’s salaries from taxation is *exactly* equivalent to paying them more – there could only be confusion about this because of the accounting trick we currently do with teacher’s salaries.
A nostril plug would not reduce carbon emissions, because twice as much carbon dioxide then has to flow out your other nostril.
Clearly I’m the only one who zeroed in on the main point of this post, while everyone else is going off on tangents.
@Steve_Landsburg: Please quote the specific portion of your post that reveals correct understanding of the issue I corrected you on.
Yes, I know it’s a parody, but it’s one predicted on a misunderstanding that you either believe or endorsed.
It’s okay to be wrong — you don’t need to make every little goof-up your “hill to die on”.
@Bennett_Haselton: A nostril plug would not reduce carbon emissions, because twice as much carbon dioxide then has to flow out your other nostril.
Bzzt. It will certainly increase your rate of breathing, but it’s not guaranteed to make up for the lost respiration — and there are metabolism-slowing effects, etc. In any case, you can’t just silently assume if you’re point is that it would accomplish nothing.
Clearly I’m the only one who zeroed in on the main point of this post, while everyone else is going off on tangents.
Well, I’m sorry, but people shouldn’t promote misconceptions, especially if they had no reason to bring up that topic to begin with. For example, someone shouldn’t bring up the genome if they can’t use the term properly and doesn’t even need to be mentioned to make their point. (Know anyone like that,
Silas: I suggest that you direct your outrage to the proprietors of The Onion. You should see some of the nonsense they’re promoting *there*!
Second time: I understand it’s a parody. I understood that the first time. Even reading that way, it’s still promoting the confusion about the carbon footprint of breathing. It would be like me writing a parody that requires you to assume protectionism generates wealth in order to “get”.
(And you still weren’t aware of what I told you about breating until today. Go ahead — find one of your past writings that shows an understanding of the carbon footprint distinction between breating and burning fossil fuels.)
In his second comment, John Faben raises an interesting point. Never mind whether gov. employees are taxed, or given what is now their after-tax paycheck tax-free; what I’d like to know is whether we should include:
a. their entire before-tax salary as a component of GDP, or just their after-tax salary;
b. their entire before-tax salary as gov. expenditure, or just their after-tax salary;
c. their tax contributions as part of tax revenue.
My hunch is that it is reasonable to include their entire before-tax salary in the GDP, but only their after-tax salary should count as gov. expenditure, and their tax contributions should not count as tax revenue.
It seems to me that the tax expenditure stuff makes some sense if we’re talking about transfer payments. Except for how the money is distributed, increasing Social Security benefits by $100 billion or lowering Social Security taxes by $100 billion should hit the government’s books in the same way…and only the books: The real economy won’t be directly affected because the government isn’t consuming more.
A few more remarks about mortgage tax relief (MTR):
* OK, MTR alleviates the distortion caused by taxing interest income; but it introduces another distortion, namely, investment in owner-occupied housing is taxed at a lower rate than investment in business, education, etc; not to mention other rental housing.
* Both MTR the tax on interest income raise interest rates (per Steve’s first comment): in this respect, their effects do not cancel out, and represent another distortion.
* To rephrase Neil’s comments: investment income is not taxed, when the investment is in owner-occupied housing. That is yet another distortion, and it persists even if MTR is abolished (because, as our host pointed out, MTR only applies while you are paying a mortgage).
The MTR DOES not alleviate the distortion of taxing interest income. Taxing interest creates a distortion by causing people to save too little. Interest tax deductions do the same thing–they cause people to borrow too much. Borrowing is negative saving.
I happen to agree with Steve that we should not tax capital income (for economic efficiency reasons). I disagree with what he says here.
There are two different ways to eliminate tax on capital income–a consumption tax or an “earnings only” tax (the idea behind the so-called “flat tax”. There is also a consumed income tax alternative).
For a consumption tax to be neutral, you have to tax housing consumption. That is one of the administrative headaches of a consumption tax. Getting people to “pre-pay” a consumption tax on housing service consumption by taxing the purchase price of a house could be a problem. The beauty of the earned income tax, is you don’t need to do that. When people spend their earnings on housing consumption, whether it be rent, mortgage interest, or mortgage principal payments, it is taxed. For efficiency, you would NOT want a mortgage interest deduction for such a tax. In other words, you do not want a mortgage interest deduction whether you want to tax capital income or not. Such a deduction just makes the saving distortion worse.
Neil: You are correct that a tax on labor income is equivalent to a consumption tax, provided you really tax all consumption.
You are correct that with a pure consumption tax, you’d want to find a way to tax housing services.
Of course, since we currently have a wage tax, housing services are already (implicitly) taxed. The fact that you’ve paid (say) half your income in taxes means, in all likelihood, that you’ve got a smaller house than you would an untaxed world, which means you’re collecting fewer housing services every year than you otherwise would have. Those missing services are the tax that you’re implicitly paying.
If you’ve borrowed money to buy that house, then someone is paying tax on the interest. That’s something we’d want to eliminate. The most straightforward way to do that is to eliminate the tax on interest. A less straightforward but equivalent way to do it is to give you a mortgage interest deduction. Both policies yield the same (after tax) interest rate to both the buyer and the lender.
Snorri:
* Both MTR the tax on interest income raise interest rates (per Steve’s first comment): in this respect, their effects do not cancel out, and represent another distortion.
This is incorrect. Suppose the interest rate (absent taxes) is 10%. Because half that interest is taxed, the interest rate rises to (say) 12%, of which the seller keeps 6%. That’s a distortion.
Now introduce a mortgage interest deduction. The interest rate rises further to 20%, of which the seller keeps 10% (just as if there were no taxes). At the same time, the buyer pays an after-tax rate of 10% (just as if there were no taxes). So the mortgage deduction erases the distortion caused by the capital tax.
(It takes some work to see that the new interest rate is exactly 20%, but even if the absence of that work you can see that the mortgage deduction at least moves things in the right direction.)
Okay, I thought you were advocating for interest deduction per se. As a second best way of getting rid of the capital tax distortion it is bad because it encourages people to overspend on housing. People may spend less on housing than they would as compared to your hypothetical “no tax” economy, but they spend more on housing than they would in a non-distorting tax economy. I think it is the latter that is relevant.
Neil: Given a tax on interest income (paid by the lender) and a mortgage interest deduction for the borrower, the equilibrium after-tax interest rate for the borrower is the same as if there were no tax. (It takes a little work to prove this, and I don’t claim to have proved it here, but it’s a standard sophomore classroom exercise.) Hence no distortion.
Steve:
This is incorrect.
OK, it’s getting late over here and I am sleepy, but your rebuttal still makes sense. (I did have the feeling that something is wrong when both a tax on interest income AND its effective rebuttal push interest rates in the same direction.)
Snorri: v(I did have the feeling that something is wrong when both a tax on interest income AND its effective rebuttal push interest rates in the same direction.)
I understand your discomfort, but the important thing is that they push quantities in *opposite* directions, and it’s the quantities that we worry about distorting.
Steve: If every borrower, domestic and foreign, were given a tax deduction at the same rate that interest is taxed, you are right–there would be no intertemporal distortion, because in a closed market a tax on one side of the market is nullified by an equivalent subsidy on the other. You cannot use that argument, however, to claim that the mortgage interest deduction does not distort the housing decision. There is an inter-sectoral distortion between housing and other forms of consumption.
there should obviously be a large tax on garden genomes, unless they are also teachers.
Neil: Fair enough.
OK, let’s see if I understand the issue of mortgage tax relief (MTR) and interest rates. This also ties up with other remarks made by Neil and by myself, and your (Steve’s) replies.
Let’s assume that “the interest rate (absent taxes) is 10%. Because half that interest is taxed, the interest rate rises to (say) 12%, of which the seller keeps 6%.”
Presumably, you and most economists would prefer to get rid of the tax on interest, instead of introducing mortgage tax relief.
But what is the effect of introducing MTR?
“The interest rate rises further to 20%, of which the seller keeps 10% (just as if there were no taxes). At the same time, the buyer pays an after-tax rate of 10% (just as if there were no taxes).”
That’s fine for the lenders and mortgage payers, but what about other borrowers? now they have to pay 20% interest, instead of 10% (without tax on interest) or 12% (with tax on interest). So while the distortion is eliminated in the mortgage business, it is increased in other areas of the economy. House prices rise; the stockmarket falls.
OK so far?
Now let’s consider homeowners who have paid off their mortgages. Can they get a second mortgage and benefit from MTR? If they do, they have an advantage in borrowing over people who do not own their houses outright: people with a 2nd mortgage effectively pay 10% interest, while the rest pays 20%.
If they don’t, then their capital is locked up in their houses, while it could be employed more profitably elsewhere, e.g. in starting a business. Interest rates suffer a further increase, because it is more difficult for borrowers to raise funds. House prices rise further; the stockmarket falls further.
Did I say something wrong?
Bennett-
In fact, the failure of the nostril plug to decrease carbon emissions is compounded by “the other nose” that gets punched, and by inciting that punch, will actually increase carbon emissions through higher respiration rates of both the puncher and the punchee.
Steve-
“The fact that you’ve paid (say) half your income in taxes means, in all likelihood, that you’ve got a smaller house than you would in an untaxed world…”
If our homebuyer lives in an untaxed world, not just he but everyone would have higher incomes to spend on housing, driving up housing prices and leaving him with the same sized house. (Unless, of course, he’s a corporate lobbyist for the mortgage industry and uses his insider knowledge to lock a contract on the bigger house before everyone else knows the untaxed world is in the pipeline.) ;)
Am I missing something?
Steve-
Neil: “Suppose the State of California wants to attract more teachers. One, it could pay them more–more budget expenditure. Two, it could exempt their income from taxes–more “tax expenditure”.”
Steve: “This one is a real expenditure…”
Let’s reframe the propostion:
Suppose the State of California wants to encourage more X. One, it can pay for it from its budget. Two, it can provide a tax break.
Doesn’t this cover virtually all current extant examples of tax expenditures?
Many of the funny, and ridiculous, examples of “unclosed” tax expenditures we can think of involve activities we don’t want to (1) discourage (like paying taxes) or (2) to provide additional encouragement for because there are other incentives sufficient to ensure the behavior (like birthdays and breathing).
At least nominally, anyway.
Back in the real world, some tax breaks are simply give-aways to favored constituencies that are publicly justified as seeking to encourage one outcome or another. Climbing on my soapbox again, the vast majority of these tax breaks go to corporations, other businesses with lobbying clout, or high-income individuals. On the rare occasion you find one of these give-aways benefitting middle or lower income individuals, it’s usually an unintended byproduct of actions taken by others in their own interest.
Of course, as with all generalizations, there are exceptions. The Per Child tax credit (refundable to low-icome taxpayers) and the Earned Income Tax Credit were not designed to encourage people to earn low incomes. Nor were they passed at the behest of business interests in service to their own interests.
But neither do they have the characteristics of a classic give-away tax break since they weren’t buried deep within impenetrable tax code provisions and passed in the dead of night. As tax expenditures they look more like government income assistance programs, except of course, they’re not transfer payments since they return income to the people who earned it.
I share Philip’s frustration with corporations lobbying for special provisions in tax codes, which are usually either well crafted loopholes or explicit favors for their business. A complete repeal of the horrendously inefficient corporate taxes would do away with that, and it would also have the “unintended” benefit of helping the poor and middle class. It would also obviate the need for corporate tax lobbyists, corporate tax accountants, and government bureaucrats who craft complicated tax policy. I believe that would cut a dead-weight loss out of our economy. Companies can charge less to their customers when their expenses are lower (due to firing the lobbyists and accountants). Also, labor is more productive when corporate profits are not taxed, so I suspect wages would inflate, too.
GregS
Yes, you’re right. I doubt that benefits to labor from eliminating all taxes on capital, if there are any net benefits, are really more than a necessary and largely unintended, but politically useful, byproduct of the disproportionate benefits they would receive for capital.
And my doubts about the net benefits to labor are *not* based on a rejection of economic theory about the flow of benefits to labor from eliminating taxes on capital. I accept these arguments. My doubts are based on crucial real-world factors that seem to be ignored in standard economic theory, such as the implied shift in tax burden to labor and the implications for increasing economic inefficiencies through permanent shifts in political power.
At a minimum, these factors suggest that benefits to labor are less than claimed by advocates of eliminating taxes on capital, if not offset, and that these benefits are being oversold.
There is no “implied shift of tax burden,” if by that you mean we must increase tax rates on labor to compensate for tax cuts on corporate profits. It’s likely we’d collect more total tax revenues if all we did was cut the corporate tax rate and did nothing else; other taxable things would happen in greater volume, so even a constant tax rate would collect higher revenues.
The tax on corporate taxes/capital is especially destructive. It turns a successful project (in an untaxed world) into an unsuccessful one in the taxed world, and it disproportionately kills projects with a low probability of a high payoff (but non-the-less positive expected profit). The projects I refer to are things like new factories, new offices, new machines, and new production lines (capital). The immediate beneficiaries of these things (capital gains) are the workers, because there are more jobs and higher paying jobs because of these things. Suppose for a moment we cut the corporate tax knowing more of these things will be built, and we expect that the budget will balance out because we’ll be taxing the new labor at the same rate as before. (This is a likely scenario, given that the cuts happen.) Is this a shift of the tax burden from capital to workers? It would be odd to describe it as such, since workers are better off and are being taxed at the same rate as before.
I’m sure you know this, but I might as well state it. You can’t simply increase or decrease tax revenue from any particular group. All you can do is change the rate of taxation; the actual effect or revenue might be in the same or the opposite direction of the change in the rate. A decrease in the corporate tax rate (or any tax rate) does not imply a “shift in tax burden” to any other group.
Again, I share your frustration about the inefficiencies due to differential political power. Design a system that has no favors to sell to corporations (such as a zero-tax, zero-subsidy system), and you’ll largely avoid this problem.
Incidentally, are you familiar with Ireland’s experiment with cutting its corporate tax rates? I’m no expert on it, and I won’t claim that example repudiates your worldview or supports mine, but I believe they credit their recent economic success with their cuts on tax rates.
“There is no “implied shift of tax burden,” if by that you mean we must increase tax rates on labor to compensate for tax cuts on corporate profits.”
Yes there is. If expenditures were constant (as is implicitly assumed in this case), the tax burden previously borne by capital *must* shift to labor, either through a rate increase or through inflation. Inflation simply delays the day the bill comes due, and since capital is tax exempt, it all falls on labor.
“It’s likely we’d collect more total tax revenues if all we did was cut the corporate tax rate and did nothing else; other taxable things would happen in greater volume, so even a constant tax rate would collect higher revenues.”
This is a *very” big leap, like supply side economics on steroids. Can you cite me a source for this conclusion?
“I’m sure you know this, but I might as well state it. You can’t simply increase or decrease tax revenue from any particular group. All you can do is change the rate of taxation; the actual effect or revenue might be in the same or the opposite direction of the change in the rate.”
I don’t see how you can come to this conclusion. This would mean all those folks who make campaign contributions and pay lobbyists to get tax laws changed are deluded in thinking they’re reducing their tax burden when in fact they’re not.
And if you believed this, why do you prefer eliminating taxes on capital instead of on labor. Wouldn’t it all wash out.
Of course there are a lot of other things we don’t tax, and the “logic” of Mr. Burman’s op-ed requires us to count all those lost opportunities as expenditures also. The Federal Government currently spends $300 billion a year on not levying a $1000 birthday tax. It spends tens of billions more on not taxing breathing
I hope you’re pulling a Schwerbitz of your own here. Because if you aren’t, your disregard of basic facts — and plain truth — is appalling.
In no way, shape or form could those ever be considered tax expenditures.
And only the Congressional Budget Office can estimate the cost to taxpayers of not taxing taxpaying.
Not the Treasury? It’s been publishing an official tax expenditure budget since 1967. Amazingly, expenditures for not taxing birthdays and breathing aren’t included in it! Because there is no “logic” at all under which they could be.
There are also several CBO papers on the subject. If you took a look at a couple you might then present the true logic of it in a way that is a bit more fair and accurate, for the benefit of your readers.
“Look. It’s really important to distinguish between spending on the one hand and not-taxing on the other. Spending consumes resources, which are then unavailable to members of the next generation. Not-taxing doesn’t do that…”
This is just plain flat wrong, when you are “not taxing” through a tax expenditure. The economic and fiscal effects of cash expenditures and tax expenditures are identical.
With a tax expenditure, just like a cash expenditure, you start with a policy baseline and then give a specially favored party a cash benefit, paid for by taxpayers, that departs from it.
For instance, the lobbyists for Purple Gizmo Makers dole out money to politicians to get a cash subsidy of $X from the Treasury.
Result: The PGMers receive $X in a transfer from taxpayers. (The taxpayers incur this cost of $X because either they must pay that much immediately to finance the subsidy, or the $X will be added to the national debt — in which case they must pay the tax cost of servicing the interest on that increase in the debt, which discounts at present value to $X. )
But wait, there’s a snag. When people hear of the subsidy check about to be written out by the Treasury, they protest: “Why gizmo manufacturers, and not all manufacturers? Why only purple gizmo manufacturers, and not us patriotic red, white and blue gizmo manufacturers? Why this spending, hand out of money?”
So the politicians go back to the drawing board and instead enact a refundable tax credit for the purple gizmo manufacturers, equal to the exact same $X.
Result: The PGMers receive $X in a transfer from taxpayers. (The taxpayers incur this cost of $X because either they must pay that much immediately to finance the subsidy, or the $X will be added to the national debt — in which case they must pay the tax cost of servicing the interest on that increase in the debt, which discounts at present value to $X. )
Identical.
Except, not politically. For those who favor small government, the tax expenditure is worse because the politicians find it so much easier to use to expand their reach over the economy while transferring more of taxpayers’ dollars to favored interests.
The politicians stand up and say “Look how we cut taxes, we’re shrinking government! We’re just letting the PGMers keep their own money to strengthen American industry, with no spending, no hand out of money! (And you buy this line, we’ll do it again and again.)”
If some voters ask, “But if you are cutting taxes why are our taxes going up? And if you are shrinking government why is the national debt growing so fast?”…
The politicians reply: “Don’t take our word for it. Here’s the noted economist Professor Schwerbitz. He’ll tell you about how special tax breaks for favored parties reduce taxes! So obviously they reduce the size of government”
“But Professor Schwerbitz, if in the first case the Purple Gizmo lobby gets $X from us taxpayers, and in the second case the Purple Gizmo lobby gets $X from us taxpayers, what’s the difference?”
“Don’t you see the obvious: in the second case the Treasury doesn’t have to write a check!”
The classic Economics in One Lesson teaches that the important thing in economics is “the thing unseen”. When protectionist policies protect jobs that are seen, the jobs that go uncreated as a result are not seen.
Here, when the cash subsidy is replaced by a tax expenditure, they fact that a check isn’t sent by the Treasury is seen — but the fact that the exact same flow of money is still going to the special interest from taxpayers nonetheless, just as if the check had been written out, is unseen.
However, it is the job of economists to see the unseen — and the ones at Treasury and OMB do see this, which is why they’ve published a detailed tax expenditure budget since the 1960s.
Now here’s another point — it is a plain contradiction from what has been claimed repeatedly elsewhere to say it is even possible to “not tax”, by deciding to not tax something, unless a matching spending cut is enacted.
If it is true that the cost of government is spending, that taxes must equal spending (and it basically is true), then one simply can’t have a tax cut without cutting spending. The supposedly new “not taxing” is only a tax shift (again a thing unseen) that must increase general tax rates. Other people must pay the shifted tax amount either immediately or later as the national debt piles up.
This is identical to the situation with a same-sized cash expenditure. If one favored party gets a money benefit of $X from taxpayers, then taxpayers must pony up $X. QED. It doesn’t matter a whit whether the favored party gets the $X by cash payment or tax break.
And with everybody’s tax rate going up we meet again our friend, the deadweight cost of taxes, which rises not with the tax rate but by the square of the increase in the tax rate … a very costly thing for us all.
Every tax expenditure can be converted into an equivalent cash subsidy. If you wouldn’t support the cash subsidy you shouldn’t support the tax expenditure. Because the cost to you — and to the entire economy through the deadweight loss cost of taxes — is the same either way.
“Crazy as all this may sound, it’s not out of the mainstream. Deficit hawks often…”
In fact, among tax and budget professionals it’s entirely mainstream among hawks and doves and just plain old non-partisan professionals alike.
Jim Glass: With a tax expenditure, just like a cash expenditure, you start with a policy baseline and then give a specially favored party a cash benefit, paid for by taxpayers, that departs from it.
So does the progressivity of the personal income tax count as a tax expenditure? Here we’re setting a top rate and then giving specially favored parties—i.e. those not in the top tax brackets—a partial exemption, which is equivalent to a cash beneift (compared to a baseline situation where everyone pays the same rate). How does that differ from any of your examples?
(Hat tip to FaTriplet3, who raised this question in the comments on Ezra Klein’s blog.)
Steve and Jim-
I worked for years as majority staff director of the Senate Finance Committee, which has jurisdiction over all taxes and govt most expenditures in the Senate, and at senior levels in Treasury, and I can attest that, with a few unimportant quibbles, Jim’s description of how “tax expenditures” are defined and treated is on the money.
And I agree with Jim that there is an important logical difference between how you are defining tax expenditures and how it is universally defined in the world of policy making, including at CBO and Treasury.
Steve, in your example, a “tax expenditure” is involved only if the current progressivity of the tax system reflects a revenue loss compared with some previous tax policy. And the tax expenditures are only the specific provisions in the change that lost revenue. The fact that the initial tax regime or the new tax regime is more or less progressive is irrelevant to whether it’s a tax expenditure.
In other words, a tax expenditure is any tax change (or a proposed one) that loses revenue from a previous (or current) or baseline level. (I don’t know if there’s a “statute of limitations” on how far back you can go in setting the baseline, but typically it goes no further back than the Tax Reform Act of 1986.) Changing the tax back in the direction of the previous baseline partially or entirely eliminates that tax expenditure.
“Tax expenditure” is never applied with respect to the potential to raise revenue by increasing a tax beyond a previous baseline or instituting an entirely new tax.
Philip:
So if the tax code used to be flat, then its current progressivity is a tax expenditure—but if it was never flat in the past, then its current progressivity is *not* a tax expenditure.
In other words, the *exact same policy* is or is not a tax expenditure depending on what policies were in place sometime in the past.
This strikes me as an extraordinarily useless concept.
Steve-
Yes, of course it’s a useless concept… if anyone were arguing for this concept. I don’t know you that might be.
You: “So if the tax code used to be flat, then its current progressivity is a tax expenditure—but if it was never flat in the past, then its current progressivity is *not* a tax expenditure.”
I don’t know where you got the impression I said anything like this. Here’s what I *did* say:
Me: “The fact that the initial tax regime or the new tax regime is more or less progressive is irrelevant to whether it’s a tax expenditure.”
What *is* relevant is whether a “tax change (or a proposed one) … loses revenue from a previous (or current) baseline level…. “Tax expenditure” is never applied with respect to the potential to raise revenue by increasing a tax beyond a previous baseline or instituting an entirely new tax.”
Where is the source of our confusion?
Steve-
“This strikes me as an extraordinarily useless concept.”
If it’s such a useless concept than how have balanced budget advocates used this concept so effectively (e.g., through devices like Gramm-Rudman) to drive us toward the budget surpluses of the Clinton years?
And how is it, now that those requirements to pay (through “offsets”) for all tax expenditures and spending proposals have expired, budget deficits have skyrocketed?
This has occurred during economic boom times when tax revenues were rising and *before” the current economic downturn drove deficits even higher.
Isn’t the implication of what you’re saying is that those budget balancers are deluded because the concept of “tax expenditure” is empty and “extraordinarily useless”?
Steve-
“So if the tax code used to be flat, then its current progressivity is a tax expenditure—but if it was never flat in the past, then its current progressivity is *not* a tax expenditure.”
I think the proper way to frame this statement so that it makes sense in terms of the concept of “tax expenditure” is…
“If the previous flat tax code is replaced by a progressive tax code that raises less revenue, then the progressive code is a tax expenditure-but if the code was never flat in the past is irrelevant because (1) the ‘flatness’ of the tax code has no bearing on whether there’s a tax expenditure, only whether there’s a loss of revenue and (2) “never… in the past” is irrelevant because only the immediate past tax code which is replaced can be the basis for calculating a tax expenditure.”
I hope this clears up the confusion.
Phillip- “If it’s such a useless concept than how have balanced budget advocates used this concept so effectively (e.g., through devices like Gramm-Rudman) to drive us toward the budget surpluses of the Clinton years?”
This is a case of “begging the question”. The context of your query presupposes that the topical “concept” caused the budget surpluses of the Clinton years.
I wasn’t a fan of Clinton at the time and my respect for his presidency has grown along with my appreciation for economics. But it was easy to promote a balanced budget plank when revenue is skyrocketing.
Spending as a % of revenue could be passed because it was relatively small but huge in terms of nominal dollars. And whoa did the Republicans laugh at Gore’s idea of a Social Security Lockbox. And whoa how they bring up Social Security’s finances as proof that a gov. run program will always fail. I get it.
But there was a lot more going on than just PAYGO that caused the budget surpluses.
My problem with Obama is that he falls into that same causal-relationship trap. The whole point of the Clinton economic plan, even Al Gore’s lock-box, was to pay down the debt during a period of economic expansion. SO when we have a period of economic contraction we will have surpluses and plenty of room under our debt ceiling which allow us to increase gov. spending.
It’s pure Larry Summers Keynesian theory and it’s brilliant. But the contract-with-America neo-con movement torpedoed the balanced budget amendment. The W. Bush years increased spending -MASSIVELY- throughout the second major period of economic expansion. and Now Obama, when an increase in gov. spending would actually help, is yielding to pressure to curtail it or to target it at the smallest number of people possible.
Krugman ran the math when he advocated a stimulus spending bill that called for 3 years of $800B spending each. Adding $2.4T to our national debt would amount to debt service fees of about 1% of our GDP. NOW is exactly the time we need to blow the debt up a bit, as long as we can have the resolve to pay it back down when the economy picks back up.
Benkyou-
Thanks for your response.
“This is a case of “begging the question”. The context of your query presupposes that the topical “concept” caused the budget surpluses of the Clinton years.”
I’m not implying it “caused” the surpluses but that it *contributed” to them and that when the legal mechanisms that operationalized the concept expired, their removel contributed to the return to deficit spending. This statement implicitly acknowledges that there were other variables at work.
“But it was easy to promote a balanced budget plank when revenue is skyrocketing.”
Yes, but Congress did not suspend the restrictions when the 1991-92 recession hit either. And it was allowed to end, and not renewed, in the boom times of the 2000s.
I agree with the rest of your post.