In my last TaxBytes column, I laid out the strange economics of so-called “tax expenditures.” I gave as an example the tax deduction for mortgage interest. That provision leads some people to own rather than rent and others to buy a more expensive house than they would have. Those are bad effects. But, I noted, it would be more straightforward for economists and politicians to say that those provisions have bad effects rather than using the convoluted language of tax expenditures.
It gets worse. Sometimes economists and politicians use the term tax expenditures to refer to tax provisions that have good effects on economic growth. They do that because they are stuck in the Haig-Simons view of income. According to this view, named after early 20th century tax scholars Robert M. Haig and Henry C. Simons, it doesn’t matter whether income comes from working or from interest and capital gains—all should be taxed equally. This view has sometimes been summarized as “A buck is a buck.”
This is from David R. Henderson, “Reducing ‘Tax Expenditures’ Can Hurt Economic Growth,” TaxBytes, Institute for Policy Innovation, September 13, 2023.
Read the whole thing.
READER COMMENTS
Vivian Darkbloom
Sep 16 2023 at 10:37am
“But, I noted, it would be more straightforward for economists and politicians to say that those provisions have bad effects rather than using the convoluted language of tax expenditures.”
No. The concept and the term “tax expenditures” has little to do with whether those items have good or bad economic effects. Rather, the term, as very ably and cogently explained by Jim Glass in the earlier thread, is used to estimate the magnitude of the actual or potential budget effect of certain, exemptions, deductions, credits, etc. There is no economic value judgement as to whether those are good or bad. But, if you want to estimate the magnitude of the bad *or* the good effects, then such an analysis is a necessary element in the process.
Also, as to the semantics of the term, as noted by Sumner in the earlier discussion, the difference between allowing a deduction, credit, etc over writing a check is often just a matter of form. Insted of giving you a tax credit, the government could simply collect the “normal” tax and then write you a check. It is for this reason that Surrey claimed many tax credits, exemptions and deductions really constitute government spending in disguise. You can view those specific provisions as good or bad economically but if you believe that most government spending is bad economically, then most tax expenditures are bad. We can have an endless debate as to what the proper “baseline” should be in calculating these “tax expenditures”; but, such discussion should not obviate the validity or usefulness of the concept.
Readers would do well to go back to the original source of the term as set out by Professor Stanley S. Surrey in 1973 (in collaboration with Paul R. McDaniel) in their book “Pathways to Tax Reform; the Concept of Tax Expenditures”. If you don’t understand the context, you won’t understand this somewhat metaphorical term (and, as *I* pointed out earlier, you also won’t understand the equally metaphorical “marginal tax rate” as often used by economists to refer to more than just tax). Surrey was concerned more about transparency than he was in judging the economic merit of any particular tax provision.
David Henderson
Sep 16 2023 at 11:51am
Vivian, a follow-up question.
As you know, high-income people face a much higher marginal tax rate than middle-income people: 37% versus 22%. Using the concept of “tax expenditures,” could one make the case that because the federal government has failed to tax the people in the 22% bracket at a rate of 37%, the revenue that would have resulted if the middle-income people had been taxed at 37% is a “tax expenditure?” And if not, why not?
I don’t ask this question as a “gotcha,” but, rather, for my own understanding. A completely acceptable answer, given that you know the Surrey article much better than I do, is that I should read that article and that it will answer my question.
Vivian Darkbloom
Sep 16 2023 at 1:23pm
Stanley Surrey’s starting point was what he viewed as a “normal tax system” (this is what I refer to above when I referred to “baseline”). He did not consider graduated rates to be outside such a “normal” baseline system, but “preferential rates” could be. Rather, he generally referred to credits, deductions and exemptions as potential tax expenditures, given generally to special groups, that have the tendency to expand and grow over time (even in his day) often as a disguised form of spending. Surrey generally believed that the Tax Code should be designed to raise income and not as a tool to advance social or economic policy. Again, we can debate endlessly about what such a baseline should look like; however, in his view (the lower end of) graduated rates were not considered “tax expenditures”, although certain “preferential rates” are. If you believe that it is impossible to arrive at an ideal baseline, you would be in good company with another giant of the time, Boris Bittker).
I think the idea was that graduated rates lack the character of “special” and also are not targeted to a narrow group of persons or interests (other than, of course, persons having in common a certain level of (taxable) income. As you will note from the quote referenced below, “tax expenditures” as calculated by the JCT are now defined by law for purposes of arriving at the “tax expenditure budget”. Feel free to argue that a tax system that is not a flat tax contains tax expenditures at the lower end of the scale, but Surrey didn’t. As is inevitable in such cases, the politics of definition can’t be avoided although I think the JCT has kept pretty much neutral, as has the CBO which does “tax expenditure” calculations all the time without referring to that term.
BTW, I referenced Surrey’s book, not an article. However, I’m sure you can find many other sources of his ideas online. This includes the preface to the Joint Committee on Taxation’s annual report on “tax expenditures” which provides a good introduction.
A brief excerpt:
“Tax expenditures are defined under the Congressional Budget and Impoundment Control Act of 1974 (the “Budget Act”) as “revenue losses attributable to provisions of the Federal tax laws which allow a special exclusion, exemption, or deduction from gross income or which provide a special credit,a preferential rate of tax,or a deferral of tax liability.” Thus, tax expenditures include any reductions in income tax liabilities that result from special tax provisions or regulations that provide tax benefits to particular taxpayers.”
I humbly submit that you are a bit too wrapped up in the semantics of the term rather than considering the entire context. As I’ve alluded to here, I had the same reaction to the term “marginal tax rate” as used by economists (including yourself on this blog multiple times). To a layperson it sounds nonsensical. If one understand the context, it’s not such a big deal (although I do think economists could have come up with a bettter, more descriptive term such as “marginal dis-incentive rate”. But, I’m not about to submit an article to the Journal of Economics complaing about it
Jim Glass
Sep 17 2023 at 1:10am
Vivian, thanks for the kind mention.
As to the rest … what you said.
David Henderson
Sep 16 2023 at 11:18am
Thanks, Vivian. I’ll check out the Surrey article.
Thomas L Hutcheson
Sep 16 2023 at 12:04pm
The “it gets worse” is just about the error of taxing income progressively instead of consumption progressively. I don’t see that is has much to do with “tax expenditure” at all.
On that, I agree with Vivian Darkbloom.
vince
Sep 16 2023 at 2:46pm
Apparently Surrey disagrees. But what if originally there was one tax rate, 37%. If Congress was considering a new 22% rate on lower incomes, wouldn’t that be a “tax expenditure”.
I believe the IRS uses related term, the revenue base, which is whatever is the current law on tax owed to the government.
He must be very disappointed.
Thomas L Hutcheson
Sep 17 2023 at 8:22am
I do not think that the proposal to create a new, lower tax bracket is usefully considered a “tax expenditure.” For me the idea of “expenditure” implies a fairly specific target.
Jim Glass
Sep 17 2023 at 12:45am
Reducing ‘Tax Expenditures’ Can Hurt Economic Growth
Keynesians and their friends have been arguing since the late 1930s that reducing government expenditures can reduce economic growth. In Europe today many are still complaining about the damage done by recent “austerity”, particularly in Britain.
Tax expenditures are just expenditures. Recipient’s assets +$X, government assets -$X, government debt increased by $X. Just like any other expenditure.
Many people have long held that reducing government expenditures can hurt economic growth.
So is this headline really news?
Thomas L Hutcheson
Sep 17 2023 at 8:26am
Reducing expenditures can reduce future income (growth) when the NPV of the expenditure >0. This is as true for governments as for firms or individuals.
Knut P. Heen
Sep 18 2023 at 5:28am
Net-present-value is an arbitrage argument. The investment costs you less today than you would receive if you sold all the cash-flows from the project in the financial markets today. You need to use exactly the correct opportunity cost of capital to calculate the net-present-value. Governments around the world tend to use an opportunity cost of capital lower than people in the private sector do. This both violates the NPV-principle, and illustrates that the return on investment is on average higher in the private sector.
Jim Glass
Sep 17 2023 at 1:02am
You keep making more of tax expenditures than they are…
Again, the tax expenditure numbers are just accounting numbers compiled in an annual accounting report. There are no “should” arguments, hidden statements about “bad effects”, “implicit assumptions” or policy opinions being expressed through them, any more than through the numbers in the cash expenditure accounting tables, the capital account tables, in the tables in the Social Security Trustees report, or in any other accounting report.
“Yo, Fred, the front office wants to know the size of the tax exemption for life insurance proceeds.”
“Give me a moment, Charlie. Checking with the Treasury…
https://home.treasury.gov/system/files/131/Tax-Expenditures-FY2023.pdf
… line 51, ‘Exclusion of life insurance death benefits’, $17,560 million this year.”
“Now they want to know how much cotton the US exported last year.”
“Aw, geeze. Over to the BEA Current Account, table 2…
https://www.bea.gov/sites/default/files/2023-03/trans422.pdf
… line 22, ‘Raw cotton’, $9,049 million in 2022″.
“HEY! Why are you guys running on with your ‘shoulds’, implicit assumptions and policy opinions hidden in all that convoluted language? Nobody asked what you think!”
“Huh?” “What?”
I really don’t believe anyone at Treasury is using that $17.56b number to imply that death benefits from life insurance shouldn’t be tax exempt, any more than the people at BEA are implying that the $9b of cotton shouldn’t have been exported. They are just accounting numbers.
Now, if you want to argue that the policies that create these numbers are good or bad or whatever, help yourself — that’s a whole ‘nother thing.
Beware tax scholars and politicians using the term “tax expenditures.”
Especially when they are quoting accountants and their accounting tables. “Who knows what evil lurks in the heart of accountancy?” 🙂
David Henderson
Sep 17 2023 at 11:05am
Good points, and I’ll hold off writing about tax expenditures until I’ve taken a careful look at the Surrey book.
I do want to point out something ironic in your hypothetical conversation, though.
Here’s what you wrote:
Notice what you didn’t write:
So even in your hypothetical, where you were completely free to choose the wording, you worded it the way I’m proposing, avoiding the “tax expenditure” language.
Ron
Sep 17 2023 at 10:24pm
It’s refreshing that you’re willing to use the moral language of “bad” effects, rather than the obfuscatory language of “tax expenditures.”
Economics as it is narrowly practiced in the U.S. is an ideology rather than some sort of normative scientific process. But admitting that would deprive economists of their raison d’etre which is to defend a specific system and debate only along the margins of this or that regulatory, social benefit scheme.
The refreshing aspect of a full throated libertarian economics is to make stated the unstated. Of course, this analysis is still not somehow scientific or objective in a meaningful way. Anymore than Marxist economic analysis within the context of figuring out how to make society better from a Marxist perspective. It just so happens that Marx was correct in a meta sense about the importance first of capitalism, and then the resolution of the capitalist system into a fully socialist one.