Re: Does Anyone Understand Why You'd Link Marginal Tax Rates To Age?

1

Why? Marginal propensity to consume in Keynesian models. Demographics, aging populations are a problem in developed economies, especially Japan, Europe and other lower immigration economies because aging populations consume less, or consume with lower multipliers.


Posted by: bob mcmanus | Link to this comment | 10-28-11 5:55 AM
horizontal rule
2

Seems like another way for baby boomers to weasel out of paying taxes.


Posted by: Spike | Link to this comment | 10-28-11 6:16 AM
horizontal rule
3
When individuals can smooth consumption privately, the level of consumption each individual controls over his or her lifetime becomes a key determinant of welfare. Therefore, the Partial Reform policy ís ability to provide higher levels of consumption to the low-skilled generates a large share of the gains from reform.

I think, think, the paper is also modeling taxing the high-income young more than the low-income young, or as a variant, letting the high-income young understand they will not be able to consume from their savings when older by taxing the savings and income of the old.

Remember, Keynes wanted virtually no private savings.


Posted by: bob mcmanus | Link to this comment | 10-28-11 6:17 AM
horizontal rule
4

2:You have it backwards. Remember, this is approved by Matt "Ummm soylent green" Yglesias, who thinks everybody over fifty is a useless racist sexist Republican.


Posted by: bob mcmanus | Link to this comment | 10-28-11 6:19 AM
horizontal rule
5

||
So last night I had a dream where I was reading a scholarly journal hundreds of years in the future. (Inevitably the device I was reading it on looked a lot like an iPad.) The issue was devoted to a bitter controversy about the correct attribution of authorship of a canonical body of work from the early 21st century. The representatives of the two sides were known as Oggedians and Wolf/sonians, respectively.
|>


Posted by: Gonerill | Link to this comment | 10-28-11 6:23 AM
horizontal rule
6

From the paper:

Why do the more sophisticated planners use lower average taxes on individuals when they are young? The data show wages rising over these age ranges in all education groups. Individuals want to borrow against future wages to raise consumption when young, but in this baseline economy they cannot transfer resources across periods. Age-dependent tax policy can sub- stitute for private borrowing by lowering average taxes when wages are low: i.e., in workers' early working years.

That's the simple model, with no borrowing/saving. In the more complicated/realistic models, borrowing and saving are permitted, but the basic effect still holds.

The basic assumption is that people want to smooth consumption over their lifetime. So, they borrow when young (when their wages are lower), payback debt and save for retirement in their prime working years, and then draw down on savings in their retirement. Age-indexed tax brackets just assist people with every step of that process--it's using the tax code to do some of that consumption smoothing for them. (Which is helpful to them, since borrowing isn't costless (when it's even possible).)


Posted by: urple | Link to this comment | 10-28-11 6:24 AM
horizontal rule
7

Of course, that means a 25 year old earning 100k pays less in taxes than a 50 year old earning 100k, which seems somewhat unfair (since, as you say, the 25 year old probably has less financial obligations than the 50 year old). But it's based on the assumption that the 25 year old will be paying much more in taxes when he or she turns 50, because his or her earnings will be much higher. (And I only skimmed the paper, but it claims teh result holds even if assumptions about the earnings path are relaxed, and individual earnings paths are assumed to be stochastic (which is of course much more realistic), as long as the basic pattern holds for the population (which is probably a fair assumption).)


Posted by: urple | Link to this comment | 10-28-11 6:30 AM
horizontal rule
8

I don't know if this is in the article, but lower tax rates on the young induce them to invest in human capital that will pay off over their entire life. If you assume hyperbolic discounting, it might encourage more rational skill acquisition.


Posted by: salacious | Link to this comment | 10-28-11 6:31 AM
horizontal rule
9

I may come back to this later, but my first thought yesterday was that under current conditions, age-dependent taxation was an unneeded complication.

We can tax savings/capital/property and make consumption/interest/borrowing deductible and achieve something close to the same effect by adjusting the rates.

Yglesias wants to end the home mortgage deduction and is attracted to the VAT and other progressive consumption taxes. He scares me.


Posted by: bob mcmanus | Link to this comment | 10-28-11 6:35 AM
horizontal rule
10

It's only three days from Halloween.


Posted by: Moby Hick | Link to this comment | 10-28-11 6:38 AM
horizontal rule
11

A lot of government transfers aren't from rich people to poor people, but rather from people at their peak income period of life to younger and older people. This includes both retirement funding and education funding. To the extent that you think retirement and education transfers are good, it stands to reason that you think people should be taxed highest during their peak income time. This should be true even for rich people.

I'm not quite sure how this interacts with having kids though. It makes more sense in say rich northern european countries where child care is government subsidized. In the U.S. where parents fund childcare and much of college, things are more complicated because at peak income is when people often have kids in college.


Posted by: Unfoggetarian: "Pause endlessly, then go in" (9) | Link to this comment | 10-28-11 6:51 AM
horizontal rule
12

5 is awesome.


Posted by: essear | Link to this comment | 10-28-11 6:51 AM
horizontal rule
13

So what happens if a 25 year old is married to a 50 year old and they want to file jointly? Do they file as a 37.5 year old?


Posted by: MAE | Link to this comment | 10-28-11 6:51 AM
horizontal rule
14

But was there really no one who argued that both ogged and wolfs/n were fictional creations of a mystery figure, balding-man-in-basement?


Posted by: essear | Link to this comment | 10-28-11 6:52 AM
horizontal rule
15

Yggles

It's not just that Social Security and Medicare are exempted from race-linked hostility to redistribution, support for Social Security and Medicare benefits is in part a kind of white racial solidarity gesture.

I assume the target of Matt's generational war isn't really the old poor black woman* totally dependent on her Social Security check but she doesn't have political power and I suspect she will be the one who gets hurt most.

*or the disabled middle-aged woman who can't get full medicare or medicaid benefits because she would like to keep her paid-off family home, as I remember from a Huffington post series of articles.


Posted by: bob mcmanus | Link to this comment | 10-28-11 6:55 AM
horizontal rule
16

Yeah, bob, the thing about Medicaid is it's specifically for people who are poor and if you have a fully paid-off family home then you are, in fact, not poor. Even if you have no other assets at all, just through owning a typical house you have more than double the net worth of the median American.


Posted by: ajay | Link to this comment | 10-28-11 7:08 AM
horizontal rule
17

What's really ghoulish about bob is that no actual individual is important to him--only abstract types. Gross. Oh. And dogs.


Posted by: Annelid Gustator | Link to this comment | 10-28-11 7:09 AM
horizontal rule
18

Does the model make any assumptions about rationality in spending habits, and is that relevant to the stimulus effect of the tax shift?

Young people tend to spend more frivolously, and are more susceptible to advertising. Old people are more practical and have very fixed spending patterns. If you gave old people more money, they might not spend it at all. If you gave young people more money, they might spend it foolishly. I don't know if they'd actually invest in things like their own education.


Posted by: rob helpy-chalk | Link to this comment | 10-28-11 7:10 AM
horizontal rule
19

18: the model does not care what young people spend money on, only that they're (on average across all people) borrowing money from their older selves to do it.


Posted by: urple | Link to this comment | 10-28-11 7:13 AM
horizontal rule
20

17: I sympathize with bob on the dog issue. For the most part, dogs are better company that humans.


Posted by: rob helpy-chalk | Link to this comment | 10-28-11 7:19 AM
horizontal rule
21

Young people tend to spend more frivolously, and are more susceptible to advertising. Old people are more practical and have very fixed spending patterns.

Not sure that this is true, actually. Advertising aimed at old people is pretty massive in US media from what I've seen - especially drugs - and I don't think they'd be making those ads if they didn't work.

Plus, older people are particularly at risk of spending their money on stupid and/or fraudulent investments, to the point where the CFPB have just named this guy http://www.housingwire.com/2011/10/19/cfpb-names-humphrey-to-head-office-of-older-americans
to make sure they don't.


Posted by: ajay | Link to this comment | 10-28-11 7:22 AM
horizontal rule
22

16: She was disabled iow, had very little income or chance of getting income, mostly free-lance writing. Some of 10-15k a year went to property taxes.

Sell the oldish house for 100k, move to an apartment that runs 10k a year how long does she stay independent? What's her quality of life?

I can't find the series. I am sure you will be pleased to know that the rich b died at 45.

17:The opposite of the truth. Those who strongly attach to individuals can become tribal and lacking caritas. I was old school Roman Catholic, remember.


Posted by: bob mcmanus | Link to this comment | 10-28-11 7:29 AM
horizontal rule
23

15.last: If you're disabled (and go through the agonizing process of qualifying for disability), you get Medicaid immediately, and Medicare after 2 years.


Posted by: Minivet | Link to this comment | 10-28-11 7:38 AM
horizontal rule
24

Not to mention if her income is that low, she may qualify for Medicaid already without getting disability. Spending down your assets to qualify for Medicaid is only for people who don't qualify based on income.


Posted by: Minivet | Link to this comment | 10-28-11 7:42 AM
horizontal rule
25

Huh. I get the shape of the consumption-smoothing argument now, but I'm not sure that it works.

First, doesn't it assume that propensity to borrow depends on the relationship of your current income to your maximum lifetime income, rather than on the absolute amount of your current income? That is, that a young person making 100K but expecting to make more later is going to borrow as if they're poorer than someone who's maxed out at 80K in midlife? This seems possible -- I'm not rejecting it as absurd -- but not something you'd want to take as a starting point (maybe the paper has a basis for it. If you don't make that assumption, then regular progressive tax rates seem to do a good enough job of consumption-smoothing; low income people will pay proportionately less taxes than high-income people.

Then, don't you have to assume that increasing a young person's discretionary income by reducing their taxes is going to reduce their propensity to borrow proportionately; that they're going to correctly assess that their discretionary income is now greater compared to what it's going to be in future, so they'll stop borrowing under the assumption that they'll be richer in midlife? This seems like a real stretch, particularly in light of the necessary first assumption.


Posted by: LizardBreath | Link to this comment | 10-28-11 7:52 AM
horizontal rule
26

doesn't it assume that propensity to borrow depends on the relationship of your current income to your maximum lifetime income, rather than on the absolute amount of your current income?

Yes, but I believe this is (on average) borne out by data.

Then, don't you have to assume that increasing a young person's discretionary income by reducing their taxes is going to reduce their propensity to borrow proportionately; that they're going to correctly assess that their discretionary income is now greater compared to what it's going to be in future, so they'll stop borrowing under the assumption that they'll be richer in midlife? This seems like a real stretch, particularly in light of the necessary first assumption.

Yes, it depends on assumptions of rationality that are probably unrealistic. The idea of lifetime income cycles (and consumer preferences for consumption smoothing across those income cycles) is well established (meaning, it's an idea that's used a lot in ecomonics; not meaning, it's necessarily well grounded). And it roughly fits a lot of aggregate data pretty well. But the idea that the patterns seen in the data are based on rational choices of consumers to smooth their consumption across phases of life, rather than just basically an artifact of certain ways in which our society is structured, is basically an assumption. As is the associated corollary, that if you change the incentives and tradeoffs faced in lifecycle consumption smoothing, people will rationally change their behaviors in response. This isn't a work of behavioral economics.


Posted by: urple | Link to this comment | 10-28-11 8:04 AM
horizontal rule
27

Or am I misunderstanding what 'smoothing consumption' means? The idea is for young people to be able to consume more compared to their future selves without costly borrowing, if I understand it correctly; I don't understand quite how that works unless there's some mechanism by which the age-linked tax structure is actually expected to reduce the costly borrowing.


Posted by: LizardBreath | Link to this comment | 10-28-11 8:05 AM
horizontal rule
28

27 crossed with 26, not in response to it.


Posted by: LizardBreath | Link to this comment | 10-28-11 8:06 AM
horizontal rule
29

27: well, that's most of it. Of course, it would also help smooth out consumption even totally in the absence of borrowing (or in the absence of any changes in borrowing), which is welfare enhancing (in this model). Not everyone can borrow, or borrow enough.


Posted by: urple | Link to this comment | 10-28-11 8:09 AM
horizontal rule
30

To state 29 more clearly: in this model, the welfare enhancing effect of age-indexed tax brackets would actually be bigger in a world without any borrowing than in the real world where borrowing is permitted. (Because borrowing already allows some consumption smoothing, which is welfare enhancing).


Posted by: urple | Link to this comment | 10-28-11 8:12 AM
horizontal rule
31

I'm not sure 30 achieved its stated goal of stating 29 more clearly, but there you have it.


Posted by: urple | Link to this comment | 10-28-11 8:13 AM
horizontal rule
32

31 was clear. Nobody can deny that.


Posted by: Moby Hick | Link to this comment | 10-28-11 8:15 AM
horizontal rule
33

The assumption that smoothing consumption over a lifetime is necessarily welfare-enhancing also seems implausible to me; I can see the existence of youthful borrowing (at a greater rate than midlife borrowing) as strong evidence that there's some appetite for such smoothing, but that doesn't rule out the possibility that it is actually welfare-enhancing, over a given lifespan, to have more of your consumption in midlife than in youth. That actually seems very plausible to me.


Posted by: LizardBreath | Link to this comment | 10-28-11 8:15 AM
horizontal rule
34

33: If you had a reasonable accounting for risk and uncertainty, I don't see consumption smoothing as obviously welfare enhancing either.


Posted by: Moby Hick | Link to this comment | 10-28-11 8:19 AM
horizontal rule
35

that doesn't rule out the possibility that it is actually welfare-enhancing, over a given lifespan, to have more of your consumption in midlife than in youth. That actually seems very plausible to me

That's actually usually the assumption--that people prefer their lifetime consumption to rise gradually over time, rather than following the same bell-like curve as their earnings.


Posted by: urple | Link to this comment | 10-28-11 8:21 AM
horizontal rule
36

Yeah, I'm remembering living in NY on temp pay, $8-10/hour in the mid '90s, having all the money I wanted and saving a fair amount. Now, I'm pathetically conservative about spending and never spent that much on having fun, but anything that took money from me, now, when I have expensive swim-teams to pay for, and sent it back to me, then, seems like it would be welfare-damaging.


Posted by: LizardBreath | Link to this comment | 10-28-11 8:23 AM
horizontal rule
37

35: Then this paper still confuses me completely.


Posted by: LizardBreath | Link to this comment | 10-28-11 8:24 AM
horizontal rule
38

36: I worked my way through college (you could do that when I was young), bought my cars with cash, took on roommates most of the time, and was broke and pretty damn miserable all the time. The Missus, instead, went deep into debt (car and student loans) and had a lot of fun in college. Given our relative prosperity now, I don't begrudge her the debt that we are repaying.


Posted by: politicalfootball | Link to this comment | 10-28-11 8:29 AM
horizontal rule
39

36: well, a few notes, (1) it's much less critical for people whose parents can do a lot of the early consumption smoothing for them (by paying for their college educations, assisting them with down payments on early cars or homes, etc.) (not indicating this was you, just noting the general fact), and (2) you probably did send quite a bit of your later income back to the earlier you, to pay for law school, although maybe you didn't send it specifically to the you who worked as a temp for a few years. This doesn't apply to all people, especially not all people for every single year of life. They're population averages, although the general overall pattern of early borrowing/midlife repayment+savings/retirement actually holds true for most people, I think.


Posted by: urple | Link to this comment | 10-28-11 8:30 AM
horizontal rule
40

That said, with the exception of my college years when I was working my ass off, I had a lot of fun as a (relatively broke) young person. A higher income is compensation for getting old, but it's not compensation enough.


Posted by: politicalfootball | Link to this comment | 10-28-11 8:32 AM
horizontal rule
41

37: I don't think you're confused. I can't believe I'm having trouble finding a good illustration of this, though--they're in every econ book. Here's a bad illustration, but it's all I can find. Just imagine that the consumption line has a gradual upward slope, instead of being perfectly flat. The picture (and the concept) still makes sense, and the same basic analysis applies.


Posted by: urple | Link to this comment | 10-28-11 8:34 AM
horizontal rule
42

I was broke and pretty damn miserable all the time.... That said ... I had a lot of fun as a (relatively broke) young person. A higher income is compensation for getting old, but it's not compensation enough.

Sort of funny to string those two together. If "pretty damn miserable" was that much better than whatever you're at now, I'm really not looking forward to getting old.


Posted by: urple | Link to this comment | 10-28-11 8:39 AM
horizontal rule
43

The assumption that smoothing consumption over a lifetime is necessarily welfare-enhancing also seems implausible to me

It is not necessarily intended to welfare-enhancing for a given individual. The point is to increase productivity, aggregate demand and help redistribution to lower income scales.


Posted by: bob mcmanus | Link to this comment | 10-28-11 8:46 AM
horizontal rule
44

I think you're misreading the paper, bob.


Posted by: urple | Link to this comment | 10-28-11 8:47 AM
horizontal rule
45

44:Despite its simplicity, age dependence generates a welfare gain equal to between 0.6 and 1.5 percent of aggregate annual consumption. The gains are due to substantial increases in both efficiency and equity. When age dependence is restricted to be Pareto-improving, the
welfare gain is nearly as large.

Marco- Economists do aggregates, not individuals.

Yeah, it is about increasing the societal marginal propensity to consume.


Posted by: bob mcmanus | Link to this comment | 10-28-11 8:53 AM
horizontal rule
46

The fact that the Unfoggetariat are having trouble wrapping their collective heads around a policy of age-indexed marginal tax rates serves to confirm my impression of it as being pretty close to the conservative stereotype of the the worst sort of liberal policy making: an overly-complex wealth-redistribution scheme thought up by ivory-tower eggheads. I mean, really, I think a flat tax is a terrible idea, but at least I can understand it.


Posted by: MAE | Link to this comment | 10-28-11 8:54 AM
horizontal rule
47

This is how macro-economists use "welfare"

There are two fundamental theorems of welfare economics. The first states that any competitive equilibrium or Walrasian equilibrium leads to a Pareto efficient allocation of resources. The second states the converse, that any efficient allocation can be sustainable by a competitive equilibrium.
...Wiki
Age dependence provides especially large welfare gains for the low-skilled, but most people obtain higher utility than they would under an age-independent policy. In fact, a simulation with the added constraint that age dependence be Pareto-improving yields nearly as large a social welfare gain as does the standard, Utilitarian-optimal age-dependent policy.
...last paragraph of linked paper

No the point is not that age-dependent tax policy would necessarily be welfare-improving for high-income young people.


Posted by: bob mcmanus | Link to this comment | 10-28-11 9:08 AM
horizontal rule
48

And in any case, I think the point is not so much to tax the young high-earner less as to tax the middle-aged relatively more.

Ok, try this. The young good-earning couple buys a house at age 25 assuming that they will have more disposable income at 55 when/because they have the mortgage paid off. If the tax system is designed so that 55+ yr olds are taxed at a higher rate compared to 25 yr olds on property/income, those 25 yr olds will understand there is little to gain from buying instead of renting, from delaying or postponing consumption. The increased consumption of the 25 yr olds will increase aggregate demand/output.

But also, and perhaps mainly, the ages 45-65 are maximum saving ages, and by taxing the middle-aged at a higher rate will also increase aggregate demand.

Not, you understand, that I agree or approve of this idea. I would rather tax wealth, and don't care if the hedge fund trader is 25 or 65. And as I said, since wealth accumulates over lifetime earnings (the house above) I think we do age-dependent taxation already. We just need to increase marginal rates.


Posted by: bob mcmanus | Link to this comment | 10-28-11 9:31 AM
horizontal rule
49

If "pretty damn miserable" was that much better than whatever you're at now,

Nonono. I was trying to distinguish between my college years - which really sucked - and the remainder of my low-income-but-not-working-80-hours-a-week youth, which was much more fun.


Posted by: politicalfootball | Link to this comment | 10-28-11 9:37 AM
horizontal rule
50

41: What's puzzling me isn't the idea that people want their consumption to slope up over time, or at least not to slope sharply downhill in old age. That makes perfect sense. The bit that gets me confused is why you would think that, given a bell-shaped income distribution over an average lifespan, and a preference for a sloping-upward consumption distribution, that transferring income from people in midlife to the same people, younger, would necessarily make anyone better off. I mean, it might, but I can't figure out why you'd be sure enough of it to treat it as an obvious goal to attain.


Posted by: LizardBreath | Link to this comment | 10-28-11 9:46 AM
horizontal rule
51

50: Well, you wouldn't just transfer it to the younger, but to the younger and the aged. Presumably (and I think this is what the paper is suggesting) marginal tax rates are lower at both ends and higher in the middle. All you're doing is having tax rates assist with some of the income smoothing that people (typically and on average) want to do.


Posted by: urple | Link to this comment | 10-28-11 10:06 AM
horizontal rule
52

Marco- Economists do aggregates, not individuals.

I choose to pretend this is a misspelling of "narco-economists", which would be awesome.

Alternately, I like to believe that the correctly spelled version of this statement is on a bumper sticker somewhere.


Posted by: F | Link to this comment | 10-28-11 10:08 AM
horizontal rule
53

51: But the actual income distribution between youth and midlife (sloping upward) is similar in shape to what seems offhand to be desirable to people generally (sloping upward). And progressive taxation does a certain amount of the same smoothing, by taxing absolutely higher income people at a higher rate than absolutely lower income people.

What's the basis for saying that people actually do want to transfer from themselves in midlife to themselves as young people more than borrowing allows them to do now? That is, wouldn't you need to know the difference between the actual lifetime income distribution and the welfare-maximizing income distribution very exactly to be sure that you were messing with it in the right direction?

(Paying for education is a weird issue -- I can't see lower marginal tax rates having any meaningful effect on most youths' ability to pay for education, because someone who's not through college yet is probably low-enough income that they're paying very little in taxes This change would have its biggest effect on high-income young people, who presumably aren't having trouble paying for education.)


Posted by: LizardBreath | Link to this comment | 10-28-11 10:15 AM
horizontal rule
54

To say that another way, I can see a welfare-enhancing argument for transferring money to low-income young people from their future selves, but I can't figure out how you could possibly do that significantly by cutting their marginal federal income tax rate. I'm not clear that I do see an obvious welfare-enhancing argument for transferring money to high-income young people from their future selves.

And come to think, mightn't this have a regressive effect generally? Person X is poor when young, and high-income when midlife. X's taxes are lowered insignificantly when young, and raised significantly in midlife. Person Y is high-income when young and in midlife. Y's taxes are lowered significantly when young, and raised significantly in midlife. It looks to me like the poorer (over her lifecycle) X is subsidizing the richer Y.


Posted by: LizardBreath | Link to this comment | 10-28-11 10:22 AM
horizontal rule
55

LB, you aren't thinking like an economist.

I'm not actually interested in defending this idea, because I think it's stupid. I was just trying to help explain the theory behind the propsal in the paper.


Posted by: urple | Link to this comment | 10-28-11 10:22 AM
horizontal rule
56

Well, yeah. I'm about to launch into "See, see, this is why I think economics is stupid." This whole paper seems to be some very plausibly valid and interesting calculation that I'm not competent to evaluate, applied to a question that looks to me to have been begged to the point of complete uselessness.


Posted by: LizardBreath | Link to this comment | 10-28-11 10:24 AM
horizontal rule
57

56: That's about as good of a definition of economics as any non-specialist needs.


Posted by: Moby Hick | Link to this comment | 10-28-11 10:25 AM
horizontal rule
58

54: High income people still borrow more in their youth than in middle age--even if they're not borrowing for school (which all but the highest income people mostly are), they're borrowing for a mortgage or for car payments, etc. (People in middle age might also have car payments, but (on average) they're saving more than they're borrowing--i.e., contributing to retirement plans more than their car payment each month. Etc.)

The paper acknowledges that effects on low income people would be less significant than on higher income people (as is true for mostly anything you're doing through the tax code). This paper doesn't generally concern itself with distributional consequences.


Posted by: urple | Link to this comment | 10-28-11 10:28 AM
horizontal rule
59

What's the basis for saying that people actually do want to transfer from themselves in midlife to themselves as young people more than borrowing allows them to do now?

We are not getting anywhere. See what you have done, urple?

It is not about individual preferences or individual welfare, anymore than the efficiency of taxing the rich depends on the self-perceived benefit to a rich person.

It is about, among other things, decreasing the aggregate propensity to save.


Posted by: bob mcmanus | Link to this comment | 10-28-11 10:29 AM
horizontal rule
60

58: But if there's no reason to think that giving them more money by reducing their taxes will make them borrow less, which there doesn't seem to be, how much they borrow doesn't enter into it, does it?


Posted by: LizardBreath | Link to this comment | 10-28-11 10:34 AM
horizontal rule
61

If you believe in life-cycle consumption smoothing (as in, believe it's something people really want, as demonstrated by their revealed preferences, as shown in aggregate population data on consumption/debt/savings patterns by age cohort), then using tax brackets to automate some of that smoothing (so people don't have to incur costs to do it themselves) makes sense. If you don't, then it makes no sense whatsoever.

The OP asked "Any thoughts on how this could possibly make sense?" and I was just trying to answer that question. You have to buy into certain widespread-in-economic-modelling-but-not-necessarily-entirely-plausible assumptions. It's not crazy, but there's nothing really about it that would compel a right-thinking person believe it was necessarily a good policy, much less an important policy.


Posted by: urple | Link to this comment | 10-28-11 10:36 AM
horizontal rule
62

But if there's no reason to think that giving them more money by reducing their taxes will make them borrow less, which there doesn't seem to be, how much they borrow doesn't enter into it, does it?

If you assume that the reason they are borrowing is consciously to move some of their future income into the present, then giving them more money now (in the form of higher taxes) in exchange for less money later (in the form of lower taxes) is basically a perfect substitute for that borrowing, which, yes, assuming rational actors is a reason to think they will borrow less. That's maybe a lot of questionable assumptions, but there you have it.



Posted by: urple | Link to this comment | 10-28-11 10:42 AM
horizontal rule
63

The bit that gets me confused is why you would think that, given a bell-shaped income distribution over an average lifespan, and a preference for a sloping-upward consumption distribution, that transferring income from people in midlife to the same people, younger, would necessarily make anyone better off. I mean, it might, but I can't figure out why you'd be sure enough of it to treat it as an obvious goal to attain.

IT INCREASES EFFICIENCY!

EFFICIENCY!!!!!


Posted by: OPINIONATED ECONOMIST | Link to this comment | 10-28-11 10:43 AM
horizontal rule
64

This is a silly proposal that will never come anywhere near a legislative body. I'm more disturbed by Saiselgy's recent implicit acceptance of the reality of economic distortions from marginal tax rates.


Posted by: Eggplant | Link to this comment | 10-28-11 10:43 AM
horizontal rule
65

This is a silly proposal that will never come anywhere near a legislative body.

This, certainly.


Posted by: LizardBreath | Link to this comment | 10-28-11 10:45 AM
horizontal rule
66

I feel like I'm trying to be helpful and it's just making you irritated with me, so I'm going to stop.


Posted by: urple | Link to this comment | 10-28-11 10:46 AM
horizontal rule
67

64 cont.: One can use something like this proposal to substitute for some progressivity in the tax code, and it has the fabulous bonus of not discouraging people to earn more, so they'll be more productive!


Posted by: Eggplant | Link to this comment | 10-28-11 10:47 AM
horizontal rule
68

66: No, I'm not irritated with you so much as with Yglesias and the original paper. This seems so far from anything that might lead to sensible policy that discussing it in a policy context strikes me as absolutely loony.


Posted by: LizardBreath | Link to this comment | 10-28-11 10:50 AM
horizontal rule
69

68: That hasn't stopped Herman Cain.


Posted by: Moby Hick | Link to this comment | 10-28-11 10:52 AM
horizontal rule
70

discussing it in a policy context strikes me as absolutely loony perfect fodder for an Unfogged comment thread


Posted by: MAE | Link to this comment | 10-28-11 10:53 AM
horizontal rule
71

66: And thank you for the explanation; I didn't understand the thinking at all before you spelled it out.


Posted by: LizardBreath | Link to this comment | 10-28-11 11:01 AM
horizontal rule
72

It's funny, my initial assumption about age-linked taxes was that, of course, you'd tax young high-income earners at a higher rate, which made sense to me. Then the economics starts.


Posted by: emdash | Link to this comment | 10-28-11 11:13 AM
horizontal rule
73

72: Well, me too -- that a high-earning young person is richer, over their lifecycle, than a similarly high-earning midlife person, and so should pay more in taxes. I still find that kind of persuasive.


Posted by: LizardBreath | Link to this comment | 10-28-11 11:21 AM
horizontal rule
74

73: but just the normal progressivity of the tax code takes care of that.


Posted by: urple | Link to this comment | 10-28-11 11:24 AM
horizontal rule
75

Depends on what you mean by 'takes care of'. The high-income kid has a lower marginal utility for their money, because over their lifecycle, there's going to be more where that came from. The high-income midlife person is doing the best they're ever going to do, so they need the money more. I think there's an argument for taxing the high-income kid more heavily.


Posted by: LizardBreath | Link to this comment | 10-28-11 11:30 AM
horizontal rule
76

And thank you for the explanation; I didn't understand the thinking at all before you spelled it out.

Seconded.

Your explanation made sense to me, which is good because when I originally saw the proposal it seemed completely nuts.


Posted by: NickS | Link to this comment | 10-28-11 11:30 AM
horizontal rule
77

Am I right in thinking that liberal econ-minded people are interested in this sort of thing because they want progressivity without distortions, so they look for wealth proxies?


Posted by: Eggplant | Link to this comment | 10-28-11 11:36 AM
horizontal rule
78

||

Do you remember that douche bag Wall Street type who put up a blog post saying that he'd take our jobs if he couldn't do his, and "Do you really think that we are incapable of teaching 3rd graders and doing landscaping?" and then everyone said, "No, I don't think you can handle third graders or do landscaping. Bring it on, fucker. And oh yeah, you are an idiot" and remember how the post was taken down because the guy who wrote it looked so obviously like a tool?

The tools at the Chicago Board of Trade thought that the other tool speaks for them. So they made his post into leaflets to drop on Occupy Chicago protestors.

Sweet Jesus H Tapdancing Christ, what assholes!

|>


Posted by: rob helpy-chalk | Link to this comment | 10-28-11 12:27 PM
horizontal rule
79
Posted by: | Link to this comment | 10-28-11 12:44 PM
horizontal rule
80

I still can't get my head around it. Lower taxes aren't a substitute for borrowing. People borrow for big things, things which cost larger proportions of their income than taxes take. Cut their taxes and they'll still have to borrow.

I'm not even sure that young people borrow more than middle aged people. Mortgages, which are the largest element of household debt, aren't taken out at 25.


Posted by: jim | Link to this comment | 10-28-11 12:46 PM
horizontal rule
81
Posted by: | Link to this comment | 10-28-11 12:58 PM
horizontal rule
82

80:Okay, try to digest this from 2.1 page 29

The logic works in the other direction as well. If old, high-skilled workers are tempted to overborrow and work less when young, the planner can make such cheating more costly by distorting the high-skilled worker ís intratemporal margin when old. The condition for the intratemporal distortion on the high-skilled old worker is analogous:

(math)

so that if the worker is tempted to borrow and mimic the low-skilled worker when young, the planner levies a positive intratemporal distortion on the older worker to raise its marginal utility of consumption and discourage that deviation. Note that it is possible to have positive distortions on the high-skilled worker in both periods.

Ps:The social welfare and redistributive results of the work are shown, for instance, in fig 2 , pg 24:"Components of Welfare Gain from Partial Reform in baseline model"


Posted by: bob mcmanus | Link to this comment | 10-28-11 1:20 PM
horizontal rule
83

The case 2 welfare gains chart is on pg 32 fig 3

Therefore, the Partial Reform policy's ability to provide higher levels of consumption to the low-skilled generates a large share of the gains from reform.
...pg 32 end of "case 2"

58:This paper doesn't generally concern itself with distributional consequences. ...urple

urple, did you read the paper?


Posted by: bob mcmanus | Link to this comment | 10-28-11 1:30 PM
horizontal rule
84

If you are looking at the charts in fig 2 and fig 3 , "types" refers to low-skilled and high-skilled workers. "Intertemporal" refers to consumption smoothing over lifetime.

Redistribution from high-skilled to low-skilled workers is the major share of the welfare gain.


Posted by: bob mcmanus | Link to this comment | 10-28-11 1:38 PM
horizontal rule
85

bob, no, I didn't read the paper, I skimmed it, but I've gone back and looked at what you're quoting and I still think you're misreading it.

More generally, to LB and others, to understand whether age-indexing income taxes is something you're going to think this is a useful policy proposal, it's worth noting one of the major problems this is supposed to be solving:

As in the previous models, high-skilled young workers are inefficiently discouraged from working by higher intratemporal distortions under an age-independent tax system.

In English: high earning (n.b. conflating that with "high skill" is grating) young workers may not be working enough because they face relatively high marginal income tax rates. If that strikes you as a realistic description of a real-world problem, then age-indexed income taxes may be for you. OTOH, if that strikes you as more or less completely disconnected from reality, then you're probably not going to like this paper.


Posted by: urple | Link to this comment | 10-28-11 1:59 PM
horizontal rule
86

In English: high earning (n.b. conflating that with "high skill" is grating) young workers may not be working enough because they face relatively high marginal income tax rates.

I'm picking at you because you seem to understand this, not because you're responsible for the reasoning in the paper, but relatively high relative to what? They've got the same marginal tax rates as anyone else with their same income. I don't follow how an argument about life-cycle income smoothing turns into an argument that young people are discouraged from working by taxes in a different manner than older people.


Posted by: LizardBreath | Link to this comment | 10-28-11 2:23 PM
horizontal rule
87

(and on your general point, yes, the whole thing sounds like nonsense, but I particularly get lost on the transition from income smoothing to allegedly distortionary effects of age-constant tax rates.)


Posted by: LizardBreath | Link to this comment | 10-28-11 2:25 PM
horizontal rule
88

relatively high relative to what?

Relative to their willingness to choose work over leisure. They've got the same marginal tax rates as anyone else with their same income, but they've got higher intertemporal marginal utility per unit of consumption than someone older with the same income*, so (the theory goes) the same marginal tax rate will discourage their work more than it will discourage the work of an older worker.

*The idea bethind consumption smoothing, which maybe I should have said earlier, is that people want the marginal utility from consumption to be equal across all periods of their life cycle. And they therefore make their lifetime savings/borrowing/consumption decisions on that basis.


Posted by: urple | Link to this comment | 10-28-11 2:40 PM
horizontal rule
89

OK. I misunderstood. It's all about intertemporality (why does the paper use "intratemporality"?); borrowing is just an analogy.

But I still can't make it work on any particular example. Young man makes $100K; in middle age he makes $100K; in old age he makes $100K (trust fund?): tax him less when he's young so his marginal utility of consumption goes up, then tax him more when he's older so his marginal utility of consumption goes down -- what has been accomplished and what does it have to do with smoothing lifetime consumption? Alright, more normal example: young: $50K, middle: $100K, old: $50K. what's accomplished by agebased differentials that isn't accomplished by regular progressivity? Another example: me. Young, essentially nothing; middle, $150K; now old, $80K. When I was young tax rates meant nothing, had no impact: I didn't pay much, if any. When flush I paid off the house quick (15 year mortgages, love 'em) so now I don't have a house payment -- which is a clear intertemporal move to smooth lifetime consumption. Taxing me more when I was flush simply because I was middle aged would have left me poorer now, since I wouldn't have been able to make that intertemporal move. Looks counterproductive.

Is there an example where the idea actually works?


Posted by: jim | Link to this comment | 10-28-11 2:44 PM
horizontal rule
90

Again, if any of this strikes you as a realistic description of the behavioral choices of people you know, you might like this policy. Otherwise, probably not.


Posted by: urple | Link to this comment | 10-28-11 2:44 PM
horizontal rule
91

Okay, skimming more, I think "high skill" is being used for "highly educated", not "high earning". That's somewhat less grating. Partial apology extended to the author.


Posted by: urple | Link to this comment | 10-28-11 2:50 PM
horizontal rule
92

bob, no, I didn't read the paper, I skimmed it, but I've gone back and looked at what you're quoting and I still think you're misreading it.

That is an insult, not an argument. Better would be to refer to the charts, fig 2 and fig 3, and the 2nd and 3rd columns in those charts and tell me what you think they mean and why they are unimportant.

Back to the topic. As I was walking...

1) Assume the central bank has wage/price inflation under lock, there will be no significant wage increases for static workers

2) Then design your tax policy to increase marginal tax rates, oh, 0.003% every day. IOW, if your income remains the same, you will taking home less after-tax money every day. 1% a year.

What effect does this have on productivity, human capital investment as mentioned above, savings, consumption across income groups and age groups?

age-dependent taxation may be not be so effective if it just catches the cash-out at 50-70.

It can create a kind of pseudo-inflationary pressure on portfolios and time-management.


Posted by: bob mcmanus | Link to this comment | 10-28-11 3:03 PM
horizontal rule
93

Don't have time to go through this thread, and someone perhaps pointed it out above, but Kremer's seminal paper on tax rates and age makes the same kind of point a lot more clearly. It helps that it uses a static and not dynamic explanation. The point is that you want to be raising taxes in the inframarginal range, where they do not discourage work effort. Marginal ranges happen lower for younger people. Of course one issue with this is that people are not easily able to adjust their effort on the margin.

The dynamic framework adds some other intertemporal issues that it looks like people were getting into above.


Posted by: PGD | Link to this comment | 10-28-11 3:12 PM
horizontal rule
94

Permanent Income Hypothesis

The permanent income hypothesis (PIH) is a theory of consumption that was developed by the American economist Milton Friedman. In its simplest form, the hypothesis states that the choices made by consumers regarding their consumption patterns are determined not by current income but by their longer-term income expectations. The key conclusion of this theory is that transitory, short-term changes in income have little effect on consumer spending behavior.

As I understand it, this is a core postulate of neo-classical and New Keynesian macro-economic modeling, and the Milton Friedman variant, life-cycle model, is considered a greater achievement than his monetary theories.

It is one of, if not the main reason why the Phillips curve was abandoned. I think. Ask DeLong or Krugman if the PIH is bullshit. The math is perfect, you see.


Posted by: bob mcmanus | Link to this comment | 10-28-11 3:15 PM
horizontal rule
95

90: Yeah, that's probably a line of argument that leads to this being sort of useless to discuss any further. I'm kind of boggling at Yglesias tossing it out as an obviously good idea.


Posted by: LizardBreath | Link to this comment | 10-28-11 3:19 PM
horizontal rule
96

90:

But there ought to be some some story that can be told in support of the theory: a type-specimen, so to speak; a narrative of a particular person in a particular career progression with particular work and consumption patterns where this sort of intervention would produce "better" work and consumption patterns over that person's career and life. That I don't actually know such a person shouldn't prevent me from accepting the value of the theory, if I can be persuaded that such a person exists and is in some sense representative.


Posted by: jim | Link to this comment | 10-28-11 5:52 PM
horizontal rule
97

34

If you had a reasonable accounting for risk and uncertainty, I don't see consumption smoothing as obviously welfare enhancing either.

This follows from the decreasing marginal utility of income and is obvious in extreme cases. If you are starving one year and all your teeth fall out and then you hit the lottery you would be better off if you could send some of your winnings back in time.

Lefties generally use this sort of argument to justify income redistribution from the rich to the poor and I don't see why it doesn't apply just as well for redistribution from your richer years to your poorer years.


Posted by: James B. Shearer | Link to this comment | 10-28-11 7:30 PM
horizontal rule
98

I could buy an argument like that, but this doesn't actually work to redistribute from people when they're rich to the same people when they're poor; poor young people hardly pay any income tax, so lowering their rates doesn't have much of an effect on them. The big effect, if I'm understanding it correctly, is that fairly high-income young people will have lower taxes compared to their even higher-income older selves, and that doesn't appear to me to do anyone much good.


Posted by: LizardBreath | Link to this comment | 10-28-11 7:35 PM
horizontal rule
99

97.last: Risk and uncertainty are greater when you are redistributing on the basis of what might happen 20 years down the road. The rest of society is saying, "this rich guy can pay us later" and runs a risk of not being paid for a variety of reasons (the guy becomes poor, the guy becomes so rich he lives on interest, the tax laws change, etc).


Posted by: Moby Hick | Link to this comment | 10-28-11 7:42 PM
horizontal rule
100

Plus, I've kind of had my fill of young, high earners and they piss me off most of the time.


Posted by: Moby Hick | Link to this comment | 10-28-11 7:44 PM
horizontal rule
101

97.2: Of course, a plain progressive tax does a better job of income redistribution than this intertemporal gimmickry because it targets, you know, income instead of a poor proxy for income, age. The mythical advantage this has is that people can't choose their age and so it doesn't change their incentives to work like high marginal rates supposedly do.


Posted by: Eggplant | Link to this comment | 10-28-11 7:46 PM
horizontal rule
102

For LB.


Posted by: Von Wafer | Link to this comment | 10-28-11 8:11 PM
horizontal rule
103

101: The mechanism still uses marginal rates, though, right?


Posted by: Benquo | Link to this comment | 11- 1-11 8:14 AM
horizontal rule
104

103: I don't know; I don't read papers. I think it can, but the goal is to reduce reliance on them.


Posted by: Eggplant | Link to this comment | 11- 1-11 8:19 AM
horizontal rule
105

I wish there was the political will in the Democratic party to push for a fair, progressive tax system. E.g. No deductions of any kind, just a straight up 1% on anything over 15K, 2% on anything over 20K, etc. Then, with the excess revenue this would generate, you could address social concerns directly, by, say, providing everyone with decent, government-sponsored health care like they do in normal countries. Also a mincome and get rid of the military.


Posted by: Natilo Paennim | Link to this comment | 11- 1-11 8:44 AM
horizontal rule
106

105: actually you could provide everyone in the US with decent government health care for the amount the US government spends on health care right now. Everyone could get NHS (or rather VA) level health care for the current cost of VA plus Medicare plus Medicaid. Government run health care is staggeringly cost-effective.


Posted by: ajay | Link to this comment | 11- 1-11 9:28 AM
horizontal rule
107

Well, sure, you could do that too. But there's also the infrastructure.


Posted by: Natilo Paennim | Link to this comment | 11- 1-11 9:32 AM
horizontal rule
108

Well, IIRC if you switched to universal government-provided health care in the US, not only would you not have to raise taxes, but you'd actually have about $60 billion or so a year left over. So you could spend that on infrastructure. Or whatever.

You'd probably need to spend a lot of it on building more public parks, because you'd all be living three years longer on average, and old folk like to sit out in parks.


Posted by: ajay | Link to this comment | 11- 1-11 9:37 AM
horizontal rule
109

I wish there was the political will in the Democratic party to push for a fair, progressive tax system_________________.

Fill in the blank. There's not much political will in the Democratic party to push for a fair, progressive anything.


Posted by: urple | Link to this comment | 11- 1-11 9:45 AM
horizontal rule
110

109: Seconded.


Posted by: apostropher | Link to this comment | 11- 1-11 9:53 AM
horizontal rule
111

108: Aha! There I have you. There is no residence in the city of Mpls. that is more than 6 blocks [3/4 of a mile] from a public park.


Posted by: Natilo Paennim | Link to this comment | 11- 1-11 9:54 AM
horizontal rule
112

111: that's a proud boast.

Is that a deliberate plan by Mpls cty gvrnmnt or just the way it worked out?


Posted by: ajay | Link to this comment | 11- 1-11 9:56 AM
horizontal rule
113

Friedman killed the Phillips curve with the expectation-augmented Phillips curve. The Phillips curve was an empirical regularity that higher inflation coincided with lower unemployment. One explanation of why it worked was that when inflation was high, nominal profits were high, so firms expanded output. Friedman argued that with persistent inflation firms would eventually catch on, and raise prices instead of output. This sort-of explained the 70s, where we had high inflation and high output.

The permanent income hypothesis came about to explain why Keynes' idea of a marginal propensity to consume as a function of income didn't work when you looked at a single person over their life cycle. When people have temporary windfalls, they save more than if they have a permanent increase in income.


Posted by: Walt Someguy | Link to this comment | 11- 1-11 11:48 AM
horizontal rule