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

EconomistMom.com

David Leonhardt’s “Club Wagner”

July 13th, 2009 . by economistmom

300px-adolph_wagner

Bruce Bartlett recently called my attention to this new club he thinks I should be a member of–the NY Times’ David Leonhardt’s “Club Wagner.” As David explains on his Economix Blog:

It’s a (fictional) organization of people willing to acknowledge a basic economic reality: Taxes in the United States must rise.

At their current levels, taxes are too low to cover the kind of government that Americans have made clear they want — a government that includes Medicare, Social Security, a strong military and numerous other programs.

Our club is named after Adolf Wagner, a 19th-century German economist who predicted that taxes would rise as societies became wealthier. “As people grew more affluent,” as the writer Matt Miller has explained Wagner’s Law, “they’d want more of what only government could provide — a strong military, public order, good schools and assorted welfare benefits, services that private citizens would have trouble arranging for on their own.”

Will spending also need to be cut? Yes, especially health care spending. But spending cuts won’t be enough. Taxes will also need to account for a larger part of tomorrow’s gross domestic product — just as they account for a much larger share of G.D.P. now than they did a century ago. Done right, tax increases do not have to stifle economic growth.

David lists the “charter members” of Club Wagner (including Bruce) on that July 7th post, but then he added more members the next day after Bruce submitted some fast nominations.  Bruce is obviously one of the most active members of the (clearly multi-partisan) club, but that’s because they haven’t added me yet.  Anyone familiar with my work and obsessions over the past, oh, maybe 8-9 years knows I’ve been an (unofficial) card-carrying member of the (fictitious) club for years.  For example, on tax day 2006 (while I was working at the Brookings Institution) I took the opportunity in a Boston Globe op-ed to…. complain about the Bush tax cuts (what a surprise!).  But in that column I also reminded readers that taxes pay for things that we value and that we might even need more than tax cuts:

…[E]ven with our imperfect tax system, the revenues provided by taxes strengthen, not weaken, our nation’s economy. They fund essential public goods and services, they contribute positively to national saving, and many of the things that they fund — from highways and schools to biomedical research and national parks — indirectly create private wealth as well. As Justice Oliver Wendell Holmes put it in 1927, ”Taxes are what we pay for a civilized society.”

And I’ve always thought that if taxes are indeed what we pay for a “civilized society,” then as our economy becomes “richer” over time, we should be willing (and in fact, eager) to pay for a more civilized society.  No?

31 Responses to “David Leonhardt’s “Club Wagner””

  1. comment number 1 by: Brooks

    Re: the statement “taxes must rise”, I think some distinctions may be useful. I assume that what is meant is that tax rates must rise, deductions reduced, and/or new types of taxes created, and I agree (and I’d add that, realistically, it’s not a matter of “if” but “when” — sooner, at a lower “cost” or later at a higher “cost”, and I think that that is the real and very important message here) . But just for clarity, “taxes rising” could alternatively mean (1) revenues increasing in real dollars due to GDP growth (growth in the tax base) even if declining or stable as a % of GDP (effective tax rate) or (2) revenues rising as a % of GDP due to bracket creep.

    As for the argument/theory that as a society’s wealth grows, it will want more government spending (either in real dollar terms or as a % of GDP), it may be valid, but a couple of causes for at least a bit of skepticism occur to me:

    As background, one element that I assume is contained in the phrase “more civilized society” is philanthropy, and yes, as wealth increases, people (individually and in aggregate) are likely to be more philanthropic, giving more in dollar terms and probably generally in terms of % of income as well.

    But the next question is why we should assume that that will or should be done through government rather than privately. And the same question applies to government spending in general, even if each person assumed that benefits of such spending (and the taxes that paid for it) did not represent a mandatory transfer of wealth (either directly or indirectly via services that benefit others much more than the individuals paying most of the taxes).

    Seems to me that the justification of government doing something is generally one or more of the following:

    1) Government can do it more efficiently, e.g., by greater economies of scale.

    2) Government can do it more fairly (e.g., courts).

    3) To avoid externalities, particularly free-riders — e.g., if we had voluntary, private payments to fund Defense or infrastructure (not counting toll roads or other user-fee-based funding), those who chose not to donate would still benefit as much as those who do, or at least benefit very substantially, and the result would be less security than most want and an unfair distribution of the financial burden — and a distribution that penalizes virtue at that!)

    4) The presumption that the majority (as theoretically represented by the political process) should impose on the minority some decision on how to use even the latters” money, either (A) based on the belief that the minority does not know what’s good for them (i.e., those in the minority are failing to see that paying a given increment in taxes is a good investment for them individually) and that it is “society’s” role to force some smart(er) financial decisions upon people (even if externalities are not involved), or, if the minority is right in thinking that paying those taxes is a bad investment for them individually (B) based on the belief that the minority is applying a deficient morality — acting excessively selfish — and that the minority is justified in imposing a superior morality on the minority (that phrasing sounds ideologically disapproving on my part, but I mean it to be neutral and merely descriptive).

    What is it about increased societal wealth (or increased wealth among some segment of society) that would make satisfying any of the conditions above more likely?

    Am I missing something in my list of criteria above?

  2. comment number 2 by: Brooks

    I would add that as a society’s wealth increases, if this wealth increase is broad-based and inclusive of those at the lower end of income or wealth, there may be less need for wealth transfers (direct or indirect), ceteris paribus, so a wealthier society may reduce government spending without causing/allowing more suffering of people in these segments. And to a limited extent, of course, we see this dynamic in our economic cycles and the changes in total spending on transfer payments such as unemployment benefits.

  3. comment number 3 by: Brooks

    I should also add that I vigorously APPLAUD Leonhardt, those in the “Club”, and certainly Diane for encouraging Americans to be realistic and responsible enough to accept that taxes will have to be increased, that taxes are going to be increased sooner or later (to reflect the priorities and values of most Americans), and that, other than at times when stimulus is net beneficial, the longer we wait for these tax increases the greater the ultimate pain will be.

    As another note I think is important, we need economists, budget experts, etc. to confront the argument that raising taxes will necessarily just generate an equal or greater amount of incremental spending rather than reducing deficits at all, ceteris paribus. This strong “don’t feed the beast” argument is not the same as the discredited strong “starve the beast” argument (by “strong” I mean the contention that a dollar reduction in revenue will cause at least a full dollar reduction in spending), because, per some using this argument, Congress’ behavioral response is different: it is less likely to cut spending in response to lower revenues than it is to increase spending in response to higher revenues. Of course, the Clinton years demonstrated that deficits can indeed come down in response to higher revenues, but it would be helpful if experts would provide more analytically-based refutations of the strong “don’t feed the beast” argument (which, by the way, quite plausibly would show a weak version to be valid — i.e., that Congress tends to spend incrementally more in response to incremental revenues, but somtimes/often substantially less than dollar for dollar)

  4. comment number 4 by: Brooks

    Why in the world would Leonhardt include Holtz-Eakin among “charter members” of “Club Wagner” on the basis (if one follows Leonhardt’s link) that Holtz-Eakin said “If you do nothing on the spending side, you’re going to raise taxes whether you’re a Republican, a Democrat, or a Martian”?? That “if-then” statement is a far cry from the unconditional assertion that is Leonhardt’s stated basis for membership in Club Wagner, ackowledgement that “Taxes in the United States must rise”.

    And as demonstrated in an exchange during a Q&A session of a speaking engagement last May, this distinction is not at all a matter of parsing: Holtz-Eakin refused to concede even that revenues will or should go up (even as a result of bracket creep — let alone through tax increases) even if we do make substantial improvements on the spending side. Instead, his response to this exact question was essentially (slightly paraphrasing at most) “We don’t know if higher revenues will be needed because we’ve never gotten serious on the spending side”. In other words, when presented with a clear opportunity to acknowledge what one must acknowledge to be legitimately eligible for membership in the Wagner Club, Holtz-Eaking explicitly refused.

  5. comment number 5 by: murf

    Definitely, no. You make the incorrect assumption that taxes have no effect on the economy becoming “richer” over time. They DO have an affect, and a profound one. Over a certain level of taxes (I believe right around 19% - sum total of all taxation), taxes become a drag on economic growth, and it is economic growth that is the sine qua non of becoming a “richer” economy. The level of taxation that is required to support the “nanny state”—which is what we are becoming—will lead to weak/no growth, which will then lead to a poorer society (which by the way is where we are headed, your kids will not be better off than you, and their kids will be even worse off then them).

    And Brooks - tl;dr.

  6. comment number 6 by: murf

    Oh…and excuse me for saying it, but David Leonhardt is an idiot! A 90% upper bracket tax rate says nothing about the overall tax rate in regards to GDP, and Clinton’s 20% rate was close to the 19% overall and took place during a period of strong economic growth, which David does not take into account at all. Tax rates can be sustained slightly higher if accompanied by economic growth. It is economic growth above all else that leads to richer economies, growth, growth, growth. We MUST sustain a path that leads to growth. Higher taxation in an already anemic environment will be a disaster.

  7. comment number 7 by: Gilleland

    “At their current levels, taxes are too low to cover the kind of government that Americans have made clear they want — a government that includes Medicare, Social Security, a strong military and numerous other programs.”

    Problem: I can’t say this for certain but I’d venture that most Americans aren’t aware that they aren’t currently paying the full price of the Government they have today via deferred cost through deficit spending and unfunded liabilities for entitlements.

    I think it remains to be seen that when the full cost of the government that Americans on aggregate want is made more clear (as is slowly happening now) then we’ll see how Americans vote — A) high taxes for same or more government or B) same / lower taxes for less government.

    My guess is that the majority of Americans will vote for B as the full cost of their desired government becomes clear.

    P.S. Not to mention the risks and impacts to individual liberty (a first principle of what it means to be an American) of bigger government. Most Americans I believe still treasure the notion and the remaining reality of individual liberty in American and will resist the unaviodable impact to it from bigger government.

  8. comment number 8 by: Brooks

    murph,

    Re:
    Brooks - tl;dr.

    Although it is indeed important to tailor one’s content to one’s target audience(s), on this blog I generally don’t dumb down my comments to acommodate folks who can’t handle more than a paragraph or who prefer their reading material oversimplified (so that they can, in turn, make oversimplified arguments themselves).

  9. comment number 9 by: Brooks

    Above was addressed to “murf”, not “murph”.

  10. comment number 10 by: Anandakos

    Gilleland,

    I think you’re confusing “most of my privileged white friends” with “most Americans”.

  11. comment number 11 by: B Davis

    murf wrote:

    Definitely, no. You make the incorrect assumption that taxes have no effect on the economy becoming “richer” over time. They DO have an affect, and a profound one. Over a certain level of taxes (I believe right around 19% - sum total of all taxation), taxes become a drag on economic growth, and it is economic growth that is the sine qua non of becoming a “richer” economy.

    I have never seen any evidence that taxes above 19% become a drag on economic growth. I assume that you are referring to 19% of GDP, close to the average level that federal tax revenue have been over the last 60 years. In addition, I assume that you are making an argument similar to that made by the article titled “You Can’t Soak the Rich” which appeared in the Wall Street Journal on May 20, 2008. That article compared the top individual tax bracket to federal tax revenue (as a percentage of GDP) and concluded that the former had no effect on the latter. In fact, that article had numerous flaws which I address at this link. If you have any other evidence that taxes above a certain level are a drag on economic growth, please provide a link to it.

    I have long felt that the default policy on tax rates should be to leave them alone. That will generally cause tax revenue to stay at the same level of GDP (ignoring things like bracket creep). This allows the benefits of any real growth in the GDP to be split among taxpayers and government services. This all works well in a steady-state society where the ratio of taxpayers to receivers of government services remain stable. The trouble is that, due to the retirement of the Boomers and the aging of society, this ratio is not stable. We collected extra FICA taxes supposedly to prepare for this event but, as everyone knows, those taxes were spent, not saved. Hence, I think that an increase in taxes in inevitable. The only question is the form and the timing of the increase.

  12. comment number 12 by: murf

    Brooks: Here you go,

    http://www.ncpa.org/pub/st215?pg=4

    http://www.investopedia.com/printable.asp?a=/articles/04/071404.asp

    http://books.google.com/books?id=3e_QMFY5GPkC&pg=PA137&lpg=PA137&dq=study+on+taxes+as+percentage+of+gdp&source=bl&ots=KmcIja20IA&sig=07pEDO–Csvvu0lMnqYSN4Dot7Y&hl=en&ei=pbtdSrzeNoKINoG6-L8C&sa=X&oi=book_result&ct=result&resnum=2

  13. comment number 13 by: Gilleland

    Anandakos,
    When I was born my parents lived in a trailer park in rural North Carolina and I was eventually raised in an average middle class environment. I don’t think that qualifies as privledged.

    I do however recognize that I “won the life lottery” in terms of the parents I have (who pulled themselves up through hard work, education, and I’m sure a little luck), being born in America and not any other country, good health, and average / maybe above average general intelligence and other environmental and inhereted qualities that have provided me an advantage in a predominately free, social democratic society.

    I understand the Rawls theory of social justice which I believe justifies a degree of tax progressively but to a limit.

    My prior point is that it would be helpful for all Americans to understand the full cost of the government they have now so that they can make an informed decision in the voting booth about how much government they are actually willing and able to pay for.

    The silver lining of this economic crisis is that it is forcing Americans to become more aware of this and to confront this issue. No one knows how this will play out but my guess is that the majority of Americans will not be willing to pay for all of the Government services that they would like to have — they’d prefer to rely more on themselves, their families, and local communities to make their way in life. I could certainly be wrong and it will be interesting to see how this plays out–I don’t it to be clear cut one way or the other.

    What are you taking issue with from my post? If you think Americans in aggregate will end up voting for A (higher taxes with same or more government) I’d be interested to understand why you think that so I can learn from you.

  14. comment number 14 by: Jim Glass

    Will spending also need to be cut? Yes.

    Well then, shouldn’t that deserve a club too? So far we have two “lets raise taxes!” clubs, The Pigou club and Club Wagner (sounds like a German resort in the Caribbean) but no “lets contain unproductive spending club” at all.

    Perhaps one needs a catchy name to get a start. Maybe the “No More Payroll Taxes Paying for Yachts Club?”

    (Or the “Cut Medicaid fraud in NY and Calif to below 40% before expanding national health care Club”? Oh ….still doesn’t seem grabby enough.)

    Our club is named after Adolf Wagner, a 19th-century German economist who predicted that taxes would rise as societies became wealthier. “As people grew more affluent,” as the writer Matt Miller has explained Wagner’s Law, “they’d want more of what only government could provide — a strong military, public order, good schools and assorted welfare benefits…

    .Well, gee, Wagner was hardly the only guy even back then to have seen that the size of government grows with wealth — others who did ranged from Gibbon to Bastiat. The difference is, they didn’t endorse it as a good course to happily follow. (See: 5th Century Rome, Louis XVI.)

    Think about it. By definition, the claim is that as people become richer they somehow need to collect more and more from other people (the government having no source of funding than people, of course). How’s that work? They said it is because as societies “advance” interest groups become more sophisticated, ingrained in the power structure, and capable of extracting resources from the polity.

    Yes, taxes then do go way up … which, by the exponentially rising force of the deadweight cost of taxes, gives huge incentives to powerful interest groups to protect themselves from them and exploit them .. which makes the tax base ever less equitable and stable … until in the last century of Rome free people were literally selling themselves into slavery to escape taxes and regulation, and at the end in France the mob brought out the guillotine. So the “natural path” to ever more government spending does not always wind through a pretty garden to a happy destination.

    Maybe we need a “Gibbon Club”? (Nah, sounds like something you’d contribute to at a zoo.)

    Anyhow, as to people as they get richer organizing to have other people pay for ever more services “only a government can provide”, just what are those services today? The cited examples…

    “The military”? As a pct of GDP military spending has been declining for decades, seems not.

    “Education”? Here in NYC the public school system now pays $20,000 per student for a 50% graduation rate, while the Catholic schools spend $5,000 for at least equal — by most studies better — results for the same-quality students. Even the Village Voice — can anyone think of a more leftward newspaper that people actually read? — is now savaging the schools’ unions and the systems’ political overseers. So while it may be that “only a government” can provide that quality education at that cost, is it really what the people want?

    “Public order”? I don’t see the cost of police and courts breakiing the national budget. Are they more than a blip?

    The real budget killer of course is ever more transfers to seniors — who indeed are the richest class of seniors who ever existed, with every succeeding generation setting a new “richest ever” record — so of course they have to collect more and more from others than all seniors before them.

    With the result being that “social insurance”, sold to the public to protect the old and infirm from undeserving poverty (”eating cat food” as many like to say), pays for ever more yachting adventures (along with the medical costs of ever more millionaires and billionaires)annually.As its growing cost threatens to bankrupt the nation. Now it may be that “only a government” can accomplish such a thing, but is it really what people want?

    Let’s imagine a “national health plan for seniors” scenario covering all, including the poor, in which Ted Kennedy in the ’60s (and Bush later with his drug plan) had said, “upon enactment, everyone will start paying a tax sufficient to cover all future actuarially expected benefits, which will keep the govt operating on sound fiscal principles”. Coulda been! Of course, with real costs confronted right up front, that program would doubtless have been far more cost-effective than what we know, as well as soundly financed.

    But instead, of course, they said: “Take! Enjoy! No cost to you, we’ll drop it all in the 2030s. Now give my party your vote!” Putting us in the fiscal mess we’re in today.

    Now really, literally, “only a government” could have done this, true enough. But is it what we want? Are we happy about it? And Club Wagner endorses all this process as good? (I’ll admit how “natural” it is.)

    Bastiat expressed a different perspective on it all: “Government as the fiction through which everybody strives to live at everybody else’s expense”.

    Maybe a Bastiat Club?

  15. comment number 15 by: Brooks

    Jim,

    I don’t think your characterization of the Pigou Club as a “Let’s raise taxes!” club is valid. Tax increases on carbon could be offset with tax cuts elsewhere, yielding no net tax increase, just a change in the basis for the revenue (one that provides incentives that better address externalities).

  16. comment number 16 by: Brooks

    Jim,

    A follow-up on my comment above: If you meant specifically Greg Mankiw’s Pigou Club, my comment doesn’t strictly apply, since he does intend it to increase overall revenues, although Mankiw’s starting point is that taxes will be increased and thus he’d like to see more of it in carbon taxation, and from that starting point my comment would apply, since his argument comes back to an assumption of a given level of revenues and advocates that carbon taxation be a larger portion. See http://gregmankiw.blogspot.com/2006/10/pigou-club-manifesto.htm

  17. comment number 17 by: Brooks

    Second try on that Mankiw link:
    http://gregmankiw.blogspot.com/2006/10/pigou-club-manifesto.html

  18. comment number 18 by: B Davis

    murf wrote:

    Brooks: Here you go,

    (three links)

    I assume that these links were meant for me since it was I who asked you to post a link to any evidence that taxes above a certain level are a drag on economic growth. In any case, I notice that two of them refer to studies done by economist Gerald W. Scully in or before 1998. I quick search via google turns up a number of more recent, opposing views. For example, following is an excerpt from a 2002 New York Times article:

    “You can make a theoretical case that high taxes impede economic growth, but it is just not supported by the evidence in the U.S. or across countries,” said William Easterly, a former World Bank economist soon to join the faculty of New York University.

    One of the most interesting research papers on the subject was done a few years ago by two economists from institutions that are hardly hotbeds of liberal protax views, the University of Chicago and the University of Rochester. Nancy L. Stokey of Chicago and Sergio Rebelo, then at Rochester but now at Northwestern, noted that income tax revenue in the United States rose to 15 percent of gross domestic product in 1942, from about 2 percent in 1913, when the tax was introduced — providing a useful natural experiment about the economic effects of taxes.

    But, they concluded, “This large rise in income tax rates produced no noticeable effect on the average growth rate of the economy.”

    Regarding the comparison of tax rates and economic growth in different countries, the article says the following:

    Relatively low-tax nations like the United States and Japan did well, they found, but so did high-tax nations in Scandinavia and elsewhere. More important, the authors contradict earlier findings that purported to show that high taxes reduced growth rates. There is no such relationship, they found; many economists now agree.

    In addition, a number of studies have found that unpaid-for tax cuts reduce economic growth. Following is an excerpt from a Center on Budget and Policy Priorities article:

    Researchers at the Joint Committee on Taxation, the Congressional Budget Office, and the Brookings Institution have all found that large unpaid-for tax cuts reduce economic growth over the long run. For example, a study by Brookings Institution economist William Gale and then-Brookings economist (now CBO director) Peter Orszag concluded that making the 2001 and 2003 tax cuts permanent without offsetting their cost would be “likely to reduce, not increase, national income over the long run.” Similarly, in a study in which it examined the economic effects of reductions in individual and corporate tax rates and an increase in the personal exemption, the Joint Committee on Taxation found, “Growth effects eventually become negative without offsetting fiscal policy [i.e. without offsets] for each of the proposals, because accumulating Federal government debt crowds out private investment.”

    Finally, I did look at the numbers myself and posted an analysis at this link. As you can see from the first graph, GDP is been pretty stable over the last 40 years. The data provides no evidence that the Reagan tax cut increased economic growth or that the Clinton tax hike decreased it.

  19. comment number 19 by: murf

    1. Perhaps Mr. Easterly needs to read some more studies since it certainly has been shown that tax rates/GDP above a certain level DO put a brake on growth, especially in an anemic environment.

    2. Income taxes at 15% of GDP are below the level of 19% and so support what I stated. Also the rise came at a time of strong economic growth.

    3. I did not argue that the Reagan tax cut increased economic growth. Nor did I argue that the Clinton tax hike decreased it.

  20. comment number 20 by: Brooks

    B Davis and murf,

    Without getting into details and specifics about the question you’re debating, I’ll offer my two cents, after taxes ;-)

    Whether higher taxes and associated higher spending reduces GDP seems (to me) highly dependent on (1) what it is being spent on vs. what those funds would be spent on if left in the private sector, and the relative ROI of each in terms of GDP, and (2) for spending for particular purposes that would occur either way (whether through taxation and government spending or if left to the private sector), the relative efficiency of doing it through government vs. through the private sector.

    [I'm going on to a second paragraph now, so I'll probably lose murf]

    Re: my #1 above, it may be that, due to the “free rider” dynamic or other reasons, there is some government spending (e.g., some “public goods”) that would generate a higher GDP ROI than the same funds would if left in private hands. Plausible examples are infrastructure, education, the court system, Defense (the latter not due mainly to stimulative effect but to providing security of assets, supply of resources, etc., without which GDP would likely suffer).

    Re: #2, there are at least theoretically — and I think in reality — some functions that government can perform more efficiently (or reliably) than the private sector, due primarily to economies of scale (e.g., administration of Social Security vs. if a number of private non-profits or businesses ran it, and in the latter case profit would be added), particularly given desirable anti-trust restrictions that limit the scale of private competitors.

    If we consider the GDP generated by businesses alone and ask if higher tax rates would likely mean lower GDP, ceteris paribus, my answer would be yes, because higher tax rates mean that a given potential investment (e.g, a new venture, new plant & equipment, R&D, etc.) is less attractive and thus some will then be foregone. But the “ceteris paribus” is where the rub is: Higher tax rates, almost by definition, don’t just mean higher tax rates, but also either higher spending, lower deficits, or some combination of the two, and these factors can potentially improve the ROI on a business investment (e.g., by providing better infrastructure; by providing secure shipping lanes in international waters and secure supply of resources such as oil; by providing rule of law via the court system; etc.)

    Of course, the net effects of all of the above generally varies by the time period one is considering, since the timing of all the effects (positive and negative) is not uniform and they don’t all coincide.

  21. comment number 21 by: B Davis

    murf wrote:

    1. Perhaps Mr. Easterly needs to read some more studies since it certainly has been shown that tax rates/GDP above a certain level DO put a brake on growth, especially in an anemic environment.

    Anytime I see the words “it certainly has been shown” in economic discussion, I become wary. In fact, I checked out the three sources that you provided. As I mentioned before, two of them refer to studies done by economist Gerald W. Scully in or before 1998. In the first one, Scully writes:

    The Growth- Maximizing Tax Rate. We use data on the real rate of growth of GDP for the 46-year period from 1950 through 1995 and on federal, state and local taxes as a share of GDP for that period.10 The resulting calculations suggest that:

    * The estimated growth-maximizing tax rate for the United States is 21 percent of GDP.

    * The overall rate of economic growth that corresponds to that tax rate is 4.8 percent.

    * Instead of the growth-maximizing tax rate of 21 percent of GDP, however, taxes were 24.2 percent of GDP in 1950 and continued to rise thereafter.

    * The actual annual economic growth rate over the 1950-95 period was 3.4 percent.

    Scully says that he used data from 1950 through 1995 and that taxes were “24.2 percent of GDP in 1950 and continued to rise thereafter”. How did he calculate the “growth-maximizing tax rate of 21 percent of GDP” when taxes were above that rate for all of the period of his data? The paper’s notes contain a few formulas and statements like “Various tests for heteroskedasticity and autocorrelation were conducted. Nothing of statistical significance was found”. I suppose we are not to worry our pretty little heads about the details of those tests.

    Unfortunately, I have found that you have to worry about the details, especially in areas where various studies disagree. As can be seen in the first graph at this link, the 10-year growth in real GDP has been pretty stable since 1950. It was slightly higher in the 10-year spans ending in the late 60s and early 70s and has dropped off slightly in the past few years. This is likewise visible in real GDP growth over full business cycles, shown in the table at the bottom of the page. It hard to imagine how Scully could have drawn significant findings from this minor variation. Since he didn’t show his work, there’s no way to know. And, because he did not show his work, the study is of little use to anyone doing serious research on the subject.

    3. I did not argue that the Reagan tax cut increased economic growth. Nor did I argue that the Clinton tax hike decreased it.

    But this is very much implied by Figure II at you first link which shows the GDP growth rate dropping off steadily from 5% to zero as the tax rate increases from 21% to 35%. In fact, I think that Scully’s economic model is far too simplistic. I agree with Brook’s statement in his prior post that one large factor in the effect of taxes on GDP growth is the manner in which those taxes are spent. I also agree with other studies that the size of any resulting deficit is also likely a large factor. As we have just seen in the recent past, an unstable financial situation is very detrimental to GDP growth. Hence, I would suggest that, it is not Mr. Easterly who needs to read some more studies but anyone who relies on simple economic models such as the one put forth by Mr. Scully.

  22. comment number 22 by: Brooks

    B Davis,

    Actually, you should have become wary as soon as you saw “ncpa.org”. That’s Pete (Pierre) du Pont’s baby. Du Pont, who isn’t an economist but plays one regularly on the WSJ opinion page, is one of the few people who hold themselves out as experts who are either shameless or obliviously ignorant enough to still insist that “tax cuts increase revenues”, that the Bush tax cuts did just that, etc.

    There’s only one thing that can be learned from the National Center for Policy Analysis and that is this: If you’re going to run a pseudo-think tank to spew baloney for partisan/ideological/self-serving reasons, given your organization a grandiose name.

  23. comment number 23 by: B Davis

    Brooks wrote:

    Actually, you should have become wary as soon as you saw “ncpa.org”. That’s Pete (Pierre) du Pont’s baby. Du Pont, who isn’t an economist but plays one regularly on the WSJ opinion page, is one of the few people who hold themselves out as experts who are either shameless or obliviously ignorant enough to still insist that “tax cuts increase revenues”, that the Bush tax cuts did just that, etc.

    I had checked this Wikipedia page which states that NCPA is a conservative think tank and has been characterized by some others as a “right wing think tank”. I hadn’t known that Pete du Pont is its current Chairman of the Board or that NCPA was his baby. Thanks for the info.

  24. comment number 24 by: Jim Glass

    A half-dozen thoughts about the discussion of taxes in these 20 comments:

    [1] There’s plenty of data showing that increasing tax rates lowers economic growth, just as logic says it should.

    For instance, as to the US, the well-noted recent Romer& Romer paper (NBER WP 13624] reports: “estimates indicate that tax increases are highly contractionary. The effects are strongly significant, highly robust…”, etc.

    Internationally, here’s a chart of tax rates to growth rates for OECD countries (scroll down a bit). And there’s lots more. Nobelist Edward Prescott is strong and insistent about how tax rates reduce employment, especially in Europe. Or just do a Google search on “taxes growth OECD countries” and see all that come up.

    [3] It’s important to know that the really serious economic damage caused by taxes comes not from the dollar cost of the tax but from its deadweight cost. A dollar of tax taken from a citizen is spent by the government, so that more or less offsets depending on how efficiently the govt spends its money. But every dollar of tax also causes a deadweight cost that is pure loss to everybody.

    For a simple example, say a common voluntary exchange (sale) produces 5 cents of gain to the parties. Then the govt imposes a new 10 cent tax on it. Now noexchanges will occur any more (because they produce an after-tax loss to the parties), so GDP shrinks, and tax revenue from the tax is zero — so the result is pure loss. Every tax imposes a deadweight cost by reducing the volume of transactions.

    The critical thing here is that deadweight cost rises not with the increase in the tax rate but exponentially, by the square of the increase in the tax rate. So if a tax goes up 10% it’s deadweight cost goes up 21%. If a tax triples from 30% to 90%, as Leonhardt thinks is a good idea, it deadweight cost goes up nine-fold.

    This is why to collect any given amount of tax revenue you want to use the lowest rate on the broadest base, and if necessary multiple low-rate taxes rather than one tax that piles up to a high rate.

    There’s no “magic rate” up to which taxes are good for growth after which they are bad. They will be “good” if the gov’t spends them effectively enough to produce a benefit that outweighs both the loss of private sector spending and the deadweight cost that results from the tax. Since the deadweight cost rises exponentially, that hurdle rises at a rapid rate!

    “An across the board increase in personal tax rates involves a deadweight loss of 76 cents per dollar of revenue, and only collects about two-thirds of the revenue implied by a ’static’ calculation.” — Feldstein, NBER WP 12201.

    The deadweight cost of taxes is never mentioned by the big-taxes like Krugman. It is by Mankiw, Boskin, Feldstein, and standard 101 public finance textboks.

    [4] It’s also important to know that it’s not taxes that matter so much as spending, because as Friedman used to say, spending determines taxes. The govt has no source of spending funds but taxes (borrowing is only deferred taxation) so if you increase spending you increase taxes. Also, if you cut taxes without cutting spending you haven’t cut taxes, you’ve only deferred taxes and added interest to them, so you don’t get most of the benefit you imagine you are. (It’s like saying “I’m saving more from my paycheck by putting more on my credit card instead!”) Even the Bush Treasury said its tax cuts wouldn’t work if not financed with spending cuts.

    Look at the tax rate/growth chart cited above and see the “spending” line running under the tax line. The Romer & Romer paper says that tax increases to close a fiscal gap, catch up with existing spending (not a “real” tax increase by Friedman’s rule), aren’t nearly as bad as tax increases that fund new spending (real tax increases per Friedman). Papers on OECD growth rates find closing fiscal gaps by increasing taxes is basically neutral, while closing gaps by cutting spending boosts growth. For instance:

    “Changes in public spending have a bigger impact than tax changes do. Particularly important are changes in the public wage bill and in government transfers …

    “The magnitude of these effects, the researchers find, is substantial…. A cut in the public wage bill of 1 percent of GDP leads to an immediate increase in the investment/GDP ratio by 0.51 percentage points, by 1.83 percentage points after two years, and by 2.77 percentage points after five years.” - -NBER Digest.

    Hey, maybe we should’ve tried that as a stimulus!

    So the next time you start to think taxes, think spending instead. And when people go on to you about taxes, tell them: “It’s spending, stupid!”

    [5] When anybody compares “Sweden” or “Scandinavia” to the US, red lights and alarms should go off. Sweden has the population of NYC. Comparing it to the US and drawing a conclusion about tax policy is like comparing New Hampshire to Europe (including Spain, Turkey, eastern Germany, etc.) and concluding that obviously Europe would do better with no income tax.

    The story of Sweden that’s always 4/5th overlooked by US liberal fans of the “Swedish model”: It became one of the richest countries in Europe following just about the most free-market, closest-to-laissez-faire policies in the world; stayed rich by being a friendly trade partner to the Nazis during WWII, keeping out of the war while all its neighbors were destroyed; after emerging from the war scratch-free as the richest nation in Europe, it adopted it’s high “socialist” social spending policies; it has been sliding down the GDP-per-capita rankings ever since; to counter the cost of high taxes Sweden has adopted just about the most pro-market governance policies anywhere: fully voucherized school system, fully privatized operation of public transport, etc. It’s practically a Cato front. That affects economic performance too!

    Yet somehow, left-side fans of the “Swedish model” never mention it includes vouchers for public schools, privatization of government services, private accounts in social security and all. Same for the “Scandinavian model”. Denmark is considered by many the most free-market nation in the world apart from its social spending-taxing.

    Which leads into…

    [5] Nobody who criticizes DuPont in WSJ as being disqualified by bias and knowing no economics should then cite Madrick in the Times. Really.

    and finally…

    [6]Jim, I don’t think your characterization of the Pigou Club as a “Let’s raise taxes!” club is valid. Tax increases on carbon could be offset with tax cuts elsewhere…

    But they never are. Look at reality. The idea of the Pigouvian tax is to tax up to the cost of the negative externality. In Europe the gas tax is unquestionably far above that amount. Is the Pigou club arguing,”Reduce the European gas tax, it is too high!”? Would anyone listen if they did?

    In the US the cap-and-trade permit auctions were equivalent to an $80 billion-a-year carbon tax. CBO said that giving 15% of the permits to carbon-emitting businesses would fully compensate them. Orszag said three months ago that giving them all to businesses, instead of keeping the revenue from them, would be the “greatest act of corporate welfare in history”, sending it right into corporate profits. What happened? (And where is Krugman on the evils of corporate welfare these days? ;-) )

    Politicians use any rationale to get and spend new revenue. “It’s virtuous to tax negative externalities? Great! Let’s do it!!” But once the tax gets gamed into the system the rationale is forgotten and the tax goes up as far as politics can push it. The federal telephone tax was created as a temporary emergency luxury tax to fund the Spanish-American War, now it’s on its third century.

    And when politicians then get a big new flow of revenue what do they do? Say: “Social Security Trust Fund”. (Or “carbon permits”.)

  25. comment number 25 by: Brooks

    Jim,

    Re:
    There’s plenty of data showing that increasing tax rates lowers economic growth, just as logic says it should.

    Perhaps you are speaking only of some range, but if we start at zero taxation and zero government spending, it’s obviously quite arguable that some spending — and thus some taxation — up to some point, to provide, say, rule of law (court system; law enforcement) or perhaps even some legitimate “public goods” that government can provide more efficiently (due to economies of scale) would result in higher GDP growth (and a higher standard of living, notwithstanding deadweight loss) than would zero government.

    So there is a point, at least in theory and perhaps when the net impact of particular programs is assessed, below which reductions of government spending and related taxation would reduce GDP and standard of living, ceteris paribus, right?

    The question then (as with the Laffer Curve) becomes one of where that tipping point is for overall spending, and for what types of spending and funded via what forms of taxation or other revenues.

  26. comment number 26 by: Jim Glass

    I note that I have a hard time counting 1,2,3,4,5,6.

    Perhaps you are speaking only of some range, but if we start at zero taxation and zero government spending, it’s obviously quite arguable that some spending — and thus some taxation — … would result in higher GDP growth (and a higher standard of living, notwithstanding deadweight loss) than would zero

    Sure. That’s why I wrote of taxes…

    “They will be ‘good’ if the gov’t spends them effectively enough to produce a benefit that outweighs both the loss of private sector spending and the deadweight cost that results from the tax. Since the deadweight cost rises exponentially, that hurdle rises at a rapid rate!”

    As to the actual real-world rate ranges that the chart and papers discuss, they are in the chart and papers.

  27. comment number 27 by: Brooks

    Jim,

    Yes, I saw that paragraph, and although I wasn’t certain, I thought your implication was that the hurdle cannot be cleared. Are you saying that it can be cleared, and agreeing that at least up to some “low” level of government spending and taxation for particular purposes, we would expect that hurdle to be cleared and thus for such spending and taxation to have a net positive impact on GDP and standard of living?

    Re: the chart, I’ve also seen charts in Matt Miller’s The Tyranny of Dead Ideas that show no correlation between average annual percent growth in real GDP per capita and average taxes as a percent of GDP, 1970 - 2004 (his source: Joel Slemrod, University of Michigan). Obviously the time periods chosen and selection of variables can paint different pictures.

    That said, I’m inclined to agree with you that, generally speaking, if we move upward from where we are in taxation and spending, GDP growth would be lower, as would average standard of living, ceteris paribus, since government is generally a less efficient allocator of resources than the “invisible hand” (hence the deadweight loss) and since taxation tends to skew incentives in ways that obstruct maximum growth.

  28. comment number 28 by: B Davis

    Jim Glass wrote:

    [1] There’s plenty of data showing that increasing tax rates lowers economic growth, just as logic says it should.

    For instance, as to the US, the well-noted recent Romer& Romer paper (NBER WP 13624] reports: “estimates indicate that tax increases are highly contractionary. The effects are strongly significant, highly robust…”, etc.

    It would be more accurate to say that there are some economic studies that suggest that certain tax increases may lower economic growth. In any event, you transposed two of the numbers in the report. The Romer & Romer paper is NBER WP 13264 and can be found at this link. Unfortunately, that link shows that it costs $5 to get a PDF copy of the report. Fortunately, however, you can find a draft of the paper at this link. As you can see, the draft has the exact same title and is dated four months before the NBER paper (March 2007 versus July 2007).

    I’m familiar with this paper and took a look at it a year or two ago. In the process, I ran across a critique of it in the April 16th, 2007 posting on Econbrowser. Following are the final two paragraphs:

    The potential contribution of such other factors of course also makes it hard to trust estimates such as those displayed above for GDP. For example, the 1981 tax cut did not arrive out of a vacuum, but instead resulted from a political process for which an important determining factor was the poor economic performance under President Carter and in particular the recession of 1979-80. If one believed, as I do, that regardless of fiscal policy, an economic recession is likely to be followed by above-average economic growth as the economy starts to recover, one would attribute at least some of the rapid growth of 1983 not to the fiscal stimulus but to the fact that the economy was recovering from a recession.

    Such factors make it difficult ever to be fully persuaded by statistical efforts like the Romers’ latest study. Nevertheless, I expect that the Romers’ new series may come to be as frequently used by academic researchers as their earlier series on exogenous monetary policy changes has been, and for the same reason– dicey though the approach may be, it is not clear what alternative we have.

    The same argument applies to the Bush tax cuts. Starting in the 2001 recession, Bush cut taxes repeatedly. As with every prior recession, we recovered and GDP growth increased during that recovery. Hence, claiming that the recovery and increased GDP growth was totally due to the tax cuts is a bit like a rooster claiming credit for the dawn. In any event, there is another critique of the Romer paper at this link.

    In addition, it is important to distinguish between types of tax increases. Following is an excerpt from page 38 of the Romer paper:

    Responses to Deficit-Driven Tax Changes. In Section III, we found that the responses of real
    GDP to the two subcategories of exogenous tax changes appear to be quite different. The response to a long-run tax increase is negative, large, and highly statistically significant. In contrast, the response to a deficit-driven tax increase is positive, though not significant.

    Hence, the Romers are saying that their study suggested that deficit-driven tax increases had the effect of INCREASING gdp. You can see a graph of the estimated increase in Figure 6b of the paper.

    The point of all this is that finding an economic study that provides evidence of some fact (at least in the opinion of some of its readers) is just the beginning of investigating an issue. Before placing any faith in such a study, I read the original study myself (or at least the conclusion and key sections if pressed for time). If possible, I crunch the numbers myself. In any event, I search for critiques of the study. In the case of the Romer paper, there are some persuasive critiques of the paper. In addition, the paper itself is not claiming that all tax increases cause a decrease in GDP.

  29. comment number 29 by: B Davis

    Following is the link to the April 16th, 2007 posting on Econbrowser which is not working in my prior post.

  30. comment number 30 by: Jim Glass

    “…In addition, the paper itself is not claiming that all tax increases cause a decrease in GDP.”

    I never said all tax increases cause a decrease in GDP. I explicitly said the opposite.

    However, let me clarify below — maybe I’ll say it yet!

    In addition, it is important to distinguish between types of tax increases.

    I explicitly did that too.

    But as to my clarification …

    A few years ago, when the Bush Administration was running deficits of about 2.5% of GDP, Brad DeLong used to beat up right-wingers who denied that deficits increased interest rates using this logic and pretty much these words:

    If you don’t believe that deficits increase interest rates you don’t believe in the law of supply-and-demand — that the government increasing demand for borrowed funds increases their cost — so you don’t believe in economics. Period. You might make a case that in a special situation this effect is diminished (such as in a slump when private borrowing is low), or the increase in interest rates is coupled with something else that’s beneficial (such as stimulative deficit spending), or the effect is small when measured — but you just cannot argue that deficits do not increase interest rates, per se, by themselves, ceteris paribus, as a matter of principle, or you do not believe in supply and demand, and thus do not believe in economics. That is all.

    It is very much the same thing with the deadweight cost of taxes, which is nothing but the law of supply-and-demand applied to markets affected by taxing instead of borrowing.

    The deadweight cost and the fact that it rises exponentially, by the square of the increase in the tax rate, is public finance 101. I gave a link to a central bank explanation of this in detail (here it is again) … and to Feldstein estimating for the US, “An across the board increase in personal tax rates involves a deadweight loss of 76 cents per dollar of revenue” (oops, omitted that link) … and to Boskin discussing it in the factors that matter for growth, under that chart that showing OECD national tax level-to-growth rates:

    “Modern academic public economics concludes that immense harm is done by high marginal tax rates…

    “The economic harm done by taxes distorting private decisions to save, invest, work, and so on goes up with the square of tax rates. Doubling tax rates quadruples the cost….

    “The marginal cost is proportional to tax rates. Hence, each dollar of additional revenues costs the economy about $1.40. When it is spent - - as legislated by Congress, adjudicated in the courts, and administered by human beings — some of it is wasted, some not narrowly targeted on the intended purpose. Perhaps $0.80 or $0.90 contributes to the intended outcome in a well-run program; only $0.30 or $0.40 in a poorly designed and administered program…”

    Now if you want to deny that tax increases slow economic growth, then like those who deny that deficit spending increases interest rates, you can plead special cases one at a time.

    When Romer & Romer say “estimates indicate that tax increases are highly contractionary. The effects are strongly significant, highly robust”, you can say they are confusing the effect of taxes with those of the business cycle or money policy or whatever … you can question what a guy like Prescott knows about European labor market … say our government programs are worth the deadweight cost of taxes they impose … argue that the estimates of the deadweight cost of Feldstein and Boskin are too high, etc. All kinds of special pleading are possible case-by-case.

    But the bottom line is this: You believe the deadweight cost of taxes exists or you don’t.

    If you do, then you must believe that tax increases do retard economic growth, per se, by themselves, ceteris paribus — and that since deadweight cost rises by the square of the increase in the tax rate, this effect must become significant as tax increases become significant, QED, as a matter of principle.

    There’s no way around it. Of course, from there you can go on to say you think a given spending program is so good it gives >$1.76 of economic value for each dollar of tax cost, or that it’s worth it to slow economic growth via the tax cost of programs that increase “social justice” in your judgment, or that you think deadweight loss is rising from a smaller base level than Feldstein says so it’s not such a big deal as all that, whatever.

    But if you actually deny that when tax rates increase, their deadweight cost increases to reduce economic growth, then you are denying that deadweight cost exists, denying that supply-and-demand applies to government tax actions (for some unknown reason), saying the public finance textbooks are wrong, and leaving the field of economics — exactly like the people who actually deny that deficits increase interest rates.

    So you decide whether you accept that the deadweight cost of taxes exists, or not, and basically all the rest follows.

    The point of all this is that finding an economic study that provides evidence of some fact (at least in the opinion of some of its readers) is just the beginning of investigating an issue.

    Actually, the established textbook principle of the deadweight cost of taxes is the beginning of the issue. Then we look at empirical studies to learn more about it in practice. Just as textbook principle tells us that deficits increase interest rates, then we look to empirical studies to see by how much in different circumstances.

    The point of all this is that finding an economic study that provides evidence of some fact (at least in the opinion of some of its readers) is just the beginning of investigating an issue. Before placing any faith in such a study, I read the original study myself…

    Well, I’m glad I gave you a bunch to poke through.

    Except remember, the reality of the cost of taxes isn’t determined just by looking at some empirical studies that one may like or not like, any more than the reality of the force of gravity is determined just by empirically looking at a rock fall, balloon rise, feather drift, etc., and averaging out some conclusion one likes.

    There is a science textbook principle of gravity that explains how all those things behave, plus a lot more — like all orbiting bodies in the heavens.

    And there is a public finance textbook principle of the deadweight cost of taxes that explains what’s talked about at the links I gave to R&R, Boskin, Feldstein, the NBER Digest, plus Prescott, and the Google papers I suggested, plus a whole lot more — such as Bruce Bartlett’s take on what killed the Roman Empire, why Europe has a VAT, and why left-center analysts at places like Brookings and the Tax Policy Center are outspokenly upset at piling up the cost of health reform entirely into new higher tax brackets.

    There’s no extraordinary explanation needed when textbook principle and data agree and confirm each other, like at the sources I linked to and in the chart from Boskin of tax burdens-to-growth rates.

    The extraordinary explanations are needed for claiming textbook principle doesn’t apply, arguing that textbook principle and data that confirm each other (as in the Boskin chart) should be disregarded anyhow, and when saying that going forward in our fiscal planning textbook principle should be disbelieved or ignored.

  31. comment number 31 by: B Davis

    Jim Glass wrote:

    [5] Nobody who criticizes DuPont in WSJ as being disqualified by bias and knowing no economics should then cite Madrick in the Times. Really.

    You’re confusing Brooks and I. I cited the Times article in my post of June 16th and Brooks criticized DuPont in his post of June 17th. I simply thanked him for the info. Also, I cited the Times article not based on Madrick’s credentials (I didn’t know who he was) but based on the assumption that a Times article would likely be accurate in its quoting of William Easterly and the Stokey-Rebelo research paper.

    In addition, the paper itself is not claiming that all tax increases cause a decrease in GDP.

    I never said all tax increases cause a decrease in GDP. I explicitly said the opposite.

    I was responding to your following statement in point[1]:

    [1] There’s plenty of data showing that increasing tax rates lowers economic growth, just as logic says it should.

    That sounds pretty unconditional. I do notice that you said the following in point [3]:

    There’s no “magic rate” up to which taxes are good for growth after which they are bad. They will be “good” if the gov’t spends them effectively enough to produce a benefit that outweighs both the loss of private sector spending and the deadweight cost that results from the tax. Since the deadweight cost rises exponentially, that hurdle rises at a rapid rate!

    That does sound conditional but anyone who didn’t read or skimmed through point [3] could have easily missed it. Hence, it seemed worth pointing out. In any event, you continued with the following example:

    For a simple example, say a common voluntary exchange (sale) produces 5 cents of gain to the parties. Then the govt imposes a new 10 cent tax on it. Now noexchanges will occur any more (because they produce an after-tax loss to the parties), so GDP shrinks, and tax revenue from the tax is zero — so the result is pure loss. Every tax imposes a deadweight cost by reducing the volume of transactions.

    I agree with this particular example. However, the effect of a tax on wages would be somewhat different in that such a tax would not create an after-tax loss. As is often pointed out, this could discourage work. However, there are factors that could counter this effect. Following is an excerpt on the effect of tax cuts from page 116 of the book “The Coming Generational Storm”, co-written by economist Laurence Kotlikoff:

    There are two competing forces at play in determining whether pretax earning rise, stay the same, or fall. On the one hand, workers may say to themselves, “Boy, now that taxes are lower, I can work less and still receive the same after tax pay. I’m going to cut back my workweek.” On the other hand, they may say, “Boy, now’s a good time to work more and earn more because taxes are lower on every extra dollar I earn”. Economists call the first of these reactions the “income effect”. They call the second reaction, the “substitution” or “incentive effect”.

    Some of the best labor economists in the country have spent their lifetimes measuring the income and substitution effects. The broad consensus of these experts is that the two effects are roughly offsetting. This means that if wage tax rates are cut by, say 15 percent, tax revenues will fall by 15 percent.

    Hence, regardless of how much a textbook principle the deadweight cost of taxes may be, there are offsetting factors that cannot be ignored. The Romer study suggests that a deficit-driven tax increase can increase GDP growth and research done by the Joint Committee on Taxation, the Congressional Budget Office, and the Brookings Institution suggest that large unpaid-for tax cuts reduce economic growth over the long run (see my July 16th post).

    In any case, I don’t think that any of this changes the basic point of Diane’s original post. That might be summed up by the following excerpt that she gives from David Leonhardt’s Economix Blog:

    Will spending also need to be cut? Yes, especially health care spending. But spending cuts won’t be enough. Taxes will also need to account for a larger part of tomorrow’s gross domestic product — just as they account for a much larger share of G.D.P. now than they did a century ago. Done right, tax increases do not have to stifle economic growth.