Huh?

Have I somehow wandered into the middle of an argument with someone else? Who said that only cranks and economic illiterates disagree with me about tax cuts? Who even said that I favored the tax cuts? Other than the capital gains cut, I don't particularly care either way. Readers asked what were the substantive arguments in favor of tax cuts; I linked to one. That doesn't mean that it's guaranteed 100% correct; only that it is a serious argument that merits being taken seriously.

Alternatively, we may be arguing about the notion that the deadweight loss of the taxes necessary to fund new spending should be taken into account. Matt seems to believe that this is some kind of fringe, crank idea that only crazy 'wingers like the head of the National Bureau of Economic Research would endorse. But since the idea of deadweight loss is not controversial, it is hardly a "minority viewpoint" that we should take it into account when contemplating new spending. It is the size of the loss that is in dispute, but while Marty Feldstein's figures are at the high end, I would hardly characterize Harvard's George F. Baker Professor of Economics as some sort of crank or fringe philosopher.

Those figures are high for my taste, but the size of the figure is irrelevant to the substantive argument, which is that if you want the government to spend money on something, you should add the deadweight loss of the tax to the direct cost of the program in order to calculate what it will really cost taxpayers. Whatever the size of that deadweight loss (and I've no doubt we could have a rousing argument about how big it is), that seems like a fairly uncontroversial thing to say.

Comments

Matthew Yglesias's posts have been rather silly. No one is claiming that only cranks opposed the tax cuts. Whatever else I might say about Joe Stiglitz or Paul Krugman, I wouldn't call them cranks when they write about economics.

There are many serious economists who opposed the Bush tax cuts. They did so for two main reasons. Some believe that a society that has a high degree of inequality is an unjust society. So these people are willing to impose higher taxes for the sake of redistribution--even if that brings lower growth and a lower average standard of living.

Others opposed the tax cuts because they were not coupled with spending cuts. These economists worry about the "crowding out" of private investment and higher real interest rates. Some also feel uneasy about the large current account deficit and see fiscal restraint as a means of addressing that.

I happen to think the tax cuts were, on balance, a good idea--both because of their incentive effects and because they will force spending restraint in the future (and are doing so already). But I don't consider everyone who disagrees with me a loon...

Also, Yglesias seems to think that the Economist is basically right-wing and pro-Bush. Has he been reading the magazine over the last few years? This is the same magazine that endorsed New Labor and John Kerry, and attacked Bush's economic policies as "reckless."

To the hard-core socialists and protectionists, the sort of people who protest against "globalization" at WTO meetings, the Economist will no doubt seem right-wing. But the writers there are significantly to the left of, say, William F. Buckley (too much to the left, for my taste, though I remain a faithful reader).

Posted by Isocrates | September 7, 2007 6:55 PM

Megan, Megan, Megan. The issue here is not "are there potential tax reductions which can be justified by sound economic reasoning of some sort?"
The issue is whether there are sound economic arguments for the specific tax policies of the Bush Administration and the Republican Party--that there are not such arguments is the thesis of Jonathan Chait's book.

And, again, there is no way to escape the existence of deadweight losses. As I have tried to get you to understand before, deadweight loss issues are something you take into account when discussing the structure of the tax code, not the level of taxes that need to be collected.

Posted by Jestak | September 7, 2007 6:56 PM

And as I've said, that's only true if you take spending as a given, a dubious position.

But thoroughly irrelevant to this argument, since the item under discussion was whether we should calculate the deadweight loss of new healthcare spending. I presume you are not now arguing that this doesn't have deadweight loss, either?

Posted by Megan McArdle | September 7, 2007 7:09 PM

This comment has been deleted for rampant trolling, and its tiresome author has been banned

Posted by HH | September 7, 2007 7:17 PM

how rich the irony of comparing tax cuts to slavery ..

Posted by neil | September 7, 2007 7:39 PM

Looks like HH hit a little too close to home. The slavery analogy was stupid, but the first part of his post sounded like it made some sense.

Posted by Harry | September 7, 2007 8:01 PM

Megan - while I'm responsive to your general argument that deadweight losses should be taken into account when considering the costs of tax increases, by the same measure the effects of increased government spending should also be taken into account. While I would not expect that such an increase would make up for the deadweight losses in overall GDP nor mitigate deadweight losses themselves (increased spending would likely be in specific sectors and/or industries, and deadweight losses more widespread), still it should be taken into account somehow. How much should it be taken into account is the billion-dollar question, and one that applies both to the effects of deadweight losses and increased government spending.

And that brings me to a final point: estimates matter. Saying deadweight losses should be taken into consideration when contemplating tax increases is mere puffery unless you can attach a level of significance to the amount of consideration you want people to give. One accomplishes this by giving an estimate, and for that reason the amount of the estimate matters much more than the simple reflection that it should be taken into consideration.

The percentage of deadweight loss provided in the Feldstein example is frightfully high, so expect some flack for it. I mean, every additional dollar of spending costs another 79 cents in growth? By that logic, the Dubya tax cuts were the best investment ever, with a 79% annual return and all (minimum!). Still waiting for my share of the returns, though...

Feldstein is admittedly a genius, but if recent history tells us anything, it's that geniuses have a way of ignoring their own track record when they start making mistakes, and Feldstein's faith in the healing power of tax cuts seems way out-of-whack with reality and has been for some time now.

Moral: When you have a good point to make, pick better examples.

Posted by James Tanner | September 7, 2007 8:16 PM

And as I've said, that's only true if you take spending as a given, a dubious position.?

The word "dubious," by itself, is no more a refutation of my point than your use, yesterday, of the word "underwhelming," by itself. An adjective isn't an argument. You have to provide some sort of reason why (and please recall I've already refuted the "starve-the-beast" argument).

But thoroughly irrelevant to this argument, since the item under discussion was whether we should calculate the deadweight loss of new healthcare spending. I presume you are not now arguing that this doesn't have deadweight loss, either?

Since you opened this post by referring back to the issue of tax cuts in general, it was far from clear that you meant this as a health care policy discussion. But to address your question, the deadweight loss of any tax increase necessary to fund a universal health care program would almost certainly be more than offset by the fact that we could reduce the share of GDP we spend on health care to a level comparable to other OECD countries (almost all of which spend less than 10% of GDP on health care as opposed to over 15% here).

Posted by Jestak | September 7, 2007 9:25 PM

I.But to address your question, the deadweight loss of any tax increase necessary to fund a universal health care program would almost certainly be more than offset by the fact that we could reduce the share of GDP we spend on health care to a level comparable to other OECD countries...-Jestak

This is what is so common here. Too many commenters seem to believe in a world without tradeoffs. Yes you can impose price controls to bring medical spending down, but that policy would have some costs, in terms of rationing and reduced quality of care for many patients...

II. As I have tried to get you to understand before, deadweight loss issues are something you take into account when discussing the structure of the tax code, not the level of taxes that need to be collected.-Jestak, again

No. When determining the level of spending a rational policymaker will try to do weigh the costs and benefits associated with that spending. And among the costs are the deadweight losses associaed with the taxes necessary to fund that spending. So the deadweight losses really will influence decisions about the level of taxation and expenditure, not just the form of the taxes.

Posted by Isocrates | September 7, 2007 9:51 PM

Jestak,

I think you (and every other advocate of nationalized health care who has made the same point) are off base here:

"The deadweight loss of any tax increase necessary to fund a universal health care program would almost certainly be more than offset by the fact that we could reduce the share of GDP we spend on health care to a level comparable to other OECD countries (almost all of which spend less than 10% of GDP on health care as opposed to over 15% here)."

Your assumption that we are going to have dramatic reductions in health care costs ignores a couple of important facts about America:

Our demographic with the highest health care costs (senior citizens) is comprised of folks who have grown accustomed to a high level of care, and won't accept European- or Canadian-style rationing.

This demographic also happens to have the most political power (e.g., the AARP).

Posted by Fred | September 7, 2007 10:02 PM

You know what's great about this blog? The non-stop self-regarding defensiveness. That never, ever, ever gets old. This awesomeness of this blog will only increase as the thickness of Megan's skin approaches zero.

Also, it's comforting to know that at any given time, I can surf on over and count on Megan--in that wounded puppy dog voice--complaining about how badly someone has misunderstood her. I think she's really finding her niche.

Posted by jw | September 7, 2007 10:15 PM

Megan_McArdle: It's *true* that only cranks disagree with the existence of the Laffer curve and deadweight losses to taxation.

And I really shouldn't need to clarify this, but: NO THAT DOESN'T MEAN THERE'S CONSENSUS ON WHERE WE ARE RELATIVE TO THE GLOBAL LAFFER CURVE MAXIMUM, JUST THAT IT HAS ONE BETWEEN ZERO AND 100 PERCENT.

Posted by Person | September 7, 2007 10:37 PM

This deadweight loss concept has the air of something that was just made up. 76 cents on the dollar, in particular, has the air of figures that were counted between 33 and 38 times, instead of just once.

I wonder if it holds both ways. Does cutting taxes by a dollar really stimulate the economy by $1.76? Of course not.

If you can illustrate it with the supply-demand curve of a one-product economy, you can make $1.76 (or whatever) appear to vanish down a black hole. But it doesn't.

Dollars go into the IRS and come out of the Treasury unscathed, until conservatives start to question their legitimacy. What a complex those poor dollars must have. They are part of the economy too!

Posted by Jalmari | September 8, 2007 12:00 AM

No. When determining the level of spending a rational policymaker will try to do weigh the costs and benefits associated with that spending. And among the costs are the deadweight losses associaed with the taxes necessary to fund that spending. So the deadweight losses really will influence decisions about the level of taxation and expenditure, not just the form of the taxes.

Isocrates, thank you for responding with an argument instead of an adjective. An argument like yours could be used to justify a policy of 1) determining that some portion of government expenditures are not cost-effective, 2) cutting those expenditures, and 3) cutting taxes by a matching amount.

That, however, is not the tax policy of the Republicans; their policy is 1) cut taxes, 2) pretend that the tax cuts will be self-financing (remember that every major contender for the Republican nomination believes this), and 3) run increased deficits when the inevitable decline in revenue occurs. And the deficits have costs of their own: an almost certain future tax increase, potential crowding out, increased debt service expenses for the government. While those costs are just as tricky to estimate as deadweight losses, it's quite likely that any reduction in deadweight loss would be more or less offset by the costs of deficits.

This is what is so common here. Too many commenters seem to believe in a world without tradeoffs. Yes you can impose price controls to bring medical spending down, but that policy would have some costs, in terms of rationing and reduced quality of care for many patients...

That's right, I have a Ph.D. in economics and I "believe in a world without tradeoffs." Not hardly, as John Wayne said in a movie I watched the other night. There are a wide variety of ways in which a universal/single payer health care system could bring costs down. Here are three:

-reduced administrative costs. Maggie Mahar has a detailed rundown on the potential for this in her book Money-Driven Medicine. One of many possibilities--Mahar reports that 5% of all health insurance premiums go to cover the costs insurance companies incur when people leave one plan and enroll with another. Guess what--with a single payer system, that doesn't happen.

-reduced drug costs due to the ability of a single buyer to offset the market power of drug makers with their own.

-the potential for reducing the 20-30% of all medical care that is unnecessary; this is what Shannon Brownlee is guest-blogging about this week over at Kevin Drum's blog.

Posted by Jestak | September 8, 2007 1:34 AM

Jestak, if you believe in tradeoffs, do you think it might be worth examining the potential trade-offs involved in reducing drug maker's market power? Furthermore, if you want to reduce the market power of drug makers, why fool around with a government purchasing regime? Why not just reduce patent protection?

Posted by Will Allen | September 8, 2007 3:19 AM

Person, please, DON'T SHOUT so much. After awhile it gets tiresome.

Jestak, basic and inherent in the single-payer system is the major fault, that it is a single-payer system. When there is no other choice, when it is a monopoly (or a realistic monopoly if any other choices are prohibitively expensive for the 'average' person) there is almost no motivation to have the single-payer system do the right things. This goes from the venal individual who looks at you arrogantly while you're patiently waiting to be registered or checked in, to the doctor who just rips through your procedure (hopefully, not a surgical one) to get on to the next, because quantity is more important than quality in single-payers.

Human nature will turn all big-government (or big corporation) situations into something that ends up being very bad for people.

Posted by falkoyn | September 8, 2007 8:01 AM

I. "That, however, is not the tax policy of the Republicans; their policy is 1) cut taxes, 2) pretend that the tax cuts will be self-financing (remember that every major contender for the Republican nomination believes this), and 3) run increased deficits when the inevitable decline in revenue occurs. And the deficits have costs of their own: an almost certain future tax increase, potential crowding out, increased debt service expenses for the government. While those costs are just as tricky to estimate as deadweight losses, it's quite likely that any reduction in deadweight loss would be more or less offset by the costs of deficits."-jerak

Yes, I agree with you for the most part. My own belief is that the positive incentive effects of the tax cuts have been larger than any negative "crowding-out" effects from the deficits (I could explain why if you like). But I admit that this is an open question on which serious economists disagree.

II. "...reduced drug costs due to the ability of a single buyer to offset the market power of drug makers with their own.-jerak

You still haven't conceded that any trade-offs are involved here. You ought to admit that what you are proposing is a form of price control. Now, I understand your argument that you wish to offset the monpoly power of the pharmaceutical companies with a kind of monopsony power for the governemnt and consumer. Fine. But admit that this will reduce the profits of the drug companies, reduce their incentive to create new drugs (which is an expensive process), and likely deprive some people in the future of life-saving medicines.

Posted by Isocrates | September 8, 2007 8:09 AM

Megan, when we still knew you as Jane, you had a great post about how people argue assuming that their opponent is the opposite of them in the area they consider most important. Perhaps a reprint of that piece on your new blog is in order as it seems like people are arguing with a McCaricature.

Posted by Josh | September 8, 2007 8:45 AM

Has anyone but me noticed that since his move the Atlantic, Matt Yglesias has substituted shrillness for analysis? I used to read him faithfully, but it is getting increasingly painful. Maybe we could call it the Krugman effect.

Posted by Ken | September 8, 2007 9:22 AM

The shame of it is Isocrates, is that there is a good chance that some very useful inquiries could be performed regarding the optimal spot on the intellectual property protection curve, between encouraging innovation, and encouraging price competition, but the political environment is such that it makes exploration of the topic very difficult. The United States Congress is a less than ideal laboratory for empirical study.

Posted by Will Allen | September 8, 2007 11:31 AM

Since Jestak has a Ph.D. in economics, perhaps he could be so kind as to correct his confused ideological brother, Jalmari, who thinks that "This deadweight loss concept has the air of something that was just made up."

Deadweight loss is a well-understood and entirely non-controversial phenomenon, and will be covered in any Econ 101 textbook. Depending on whether you talk to Paul Krugman or Greg Mankiw, you will get estimates of between 15 cents and 85 cents for every extra dollar of tax under the current tax regieme, but neither will deny that there is some loss to GDP for each dollar of increased tax, and neither will deny that there is similiarly some gain to GDP from each dollar of tax cut.

Posted by David Wright | September 8, 2007 8:16 PM

Our demographic with the highest health care costs (senior citizens) is comprised of folks who have grown accustomed to a high level of care, and won't accept European- or Canadian-style rationing.

Every senior citizen is already covered by comprehensive government health care. The overwhelming majority of them use it as their primary coverage.

Ergo, we've already soared well over this contention...several decades ago.

Posted by Starscream | September 9, 2007 8:20 AM

A tax of 100% of n produces n.
A tax of 0% of n produces 0.

How is this a proof of the Laffer curve?

Posted by Jeffrey Davis | September 9, 2007 11:54 AM

Deadweight loss is a well-understood and entirely non-controversial phenomenon, and will be covered in any Econ 101 textbook. Depending on whether you talk to Paul Krugman or Greg Mankiw, you will get estimates of between 15 cents and 85 cents for every extra dollar of tax under the current tax regieme

This is more or less correct, although, to be precise, some introductory econ texts do not cover deadweight loss, aka Harberger triangles. When I teach the subject in my micro courses, I have to give my students a special handout. But it is a basic economic concept; I studied in grad school under Arnold Harberger, one of the main developers of the concept (hence, "Harberger triangles").

Estimates of the amount of deadweight loss from tax increases do indeed vary, although the numbers I have seen are a little different from David's. Feldstein's figure of 76 cents per dollar of tax seems to be a bit of an outlier; most other estimates I've come across seem to be at 30 cents per dollar or lower, and a few modern estimates are nearly as low as Harberger's initial finding of 2.5 cents per dollar.

The issue I've repeatedly argued with Megan is whether you can use deadweight losses as an argument in favor of tax cuts in general, as opposed to in favor of restructuring the tax code. I've explained upthread, in my reply to Isocrates, why deadweight loss arguments can't really justify Bush Administration style, deficit-increasing tax cuts.

Posted by Jestak | September 9, 2007 3:24 PM

A tax of 100% of n produces n.
A tax of 0% of n produces 0.
How is this a proof of the Laffer curve?

That isn't proof of the Laffer curve; that is proof you don't understand incentives.

A tax of 100% of n produces 0, not n. How many hours do you plan to spend at work when your take home pay reads $0.00?

Posted by David Nieporent | September 9, 2007 3:40 PM

Thanks for your instructive and friendly reply, Jestak.

Your point about tax code reform is well-taken. Certainly if your main concern was the the current size of dead-weight losses, it would behove you to start by reforming the tax code (i.e. reduce dead-weight losses with no loss of revenue) rather than by cutting taxes (i.e reduce dead-weight losses by reducing revenue).

But it is a bit of an evasion (a deft evasion, but an evasion nonetheless) to claim that this means increased dead-weight losses aren't a legitimate disadvantage of the sort of national health care proposals currently on offer. None of those proposals couples health care reform with tax reform in a way that is dead-weight loss neutral. In fact, the most well-developed and comprehensive proposals (e.g. Edwards) finance themselves in the worst possible way from the point of view of dead-weight losses -- by increasing the tax rates that are currently the highest. It's true that if your main concern was reducing dead-weight losses, you would probably spend your days campaigning for tax reform, but it's also true that you be be likely to oppose other proposed programs that significantly increase dead-weight losses.

Of course, none of this proves that nationalized health-care is wrong. From a utilitiarian perspective, that depends on whether the utility that the government buys through the program is more than the utility it costs to finance it. (And that depends, in turn, on how one aggregates utilities, which is rather subjective.) But Megan is right to point out that that utility cost should not only reflect the taxes collected but also the associated dead-weight losses.

Posted by David Wright | September 9, 2007 5:05 PM

The "incentive" gag finesses Time into the mix. Among other things. Since the Laffer Curve isn't described formally -- it's a back-of-the-envelope contraption -- it can change at will. It's a rhetorical "law" not a mathematical property.

Posted by Jeffrey Davis | September 10, 2007 8:21 AM