« Jon Swift | Main | Exhuming McCarthy »

March 05, 2010


For a second there, I wasn't sure whether Seb is talking about low-income corporations or low-income persons :)

Low-income corporations deserve our sympathy and respect and ought not to have to pay taxes because of all the other contributions they make.

Low-income persons are "lucky duckies", parasitic leeches who exploit the country by working low-income jobs and not paying income tax.

I'm sure Sebastian is clear on the difference between them. (Plus, corporate persons never have abortions! So that's another way in which they're morally superior.)

Yes, although getting ALL of the conservatives standing in front of the Senate blocking unemployment was really difficult so they just had one guy do it, trying to get the Senate to pay for it. Not a bad idea to try it really, before you unblock it and allow the bill, you vote for, to pass.

This claim ignores that those same conservatives voted en bloc against pay-go shortly before that.
But the current philosophy is "I proposed this very law in [time not too long ago] but if the enemy party likes it, it must be fatally flawed, so I have to vote against it." Holy Joe* even made that claim explicitly.

*not a con (yet), just scum

Tom Tomorrow has a good comic up, about Corporate Americans.

"From what I understand this is a significant factor in why many businesses consistently show no net profit year after year. I have no real problem with that, but I don't think that at the end of 10 years of reinvestment the resulting income from capital gains should be treated differently to wage income."

Where do you get this understanding? And by 'many' do you mean "a large sounding number" or "a significant fraction of corporations"

And you realize that if they expand too fast they could lose everything, right? And then not get the 10 years of wages that an employee gets.

And I don't understand your point about S-corps at all. If an S-corp is passing income to its owner (which is what S-corps do) that income is taxed at the full personal income tax rate. S-corps aren't magically escaping the tax man by funelling money to their owners. They are doing what S-corps are designed to do, make taxes easy to file for relatively small and usually owner operated corporations. And that is ht makes the NYT quote such a crock. The included in their statistic the almost 50% of corporations which would never pay the "corporate tax" and they insinuated that this was because they were somehow avoiding taxation. But they don't. They pay personal taxes (not captial gains) every year. And that isn't a dodge, that is how it is designed. That is why the classification exists.

One good point I thought was made and seems to be been ignored: capital gains tax doesn't account for inflation. Investing 100k and selling 25 years later at $150k is not the same as investing 100k and cashing out $150k 2 years later. The short-term/long-term capital gains rates don't address this particularly well.

My proposal would be to bring capital gains taxation in line with income taxation, but have a more finely-tuned adjustment (rather than just the two rates) to account for how long you've been invested. The rate of return should matter, shouldn't it?

I also think additional income tax brackets make sense. I don't see how someone making a few hundred thousand a year should be taxed at the same rate as someone making a few million a year.

More brackets, less deductions, kill the AMT (which is what, a ~45 year old patch aimed at rich people with lots of deductions?).

And don't even get me started on the "death tax" issue. What a joke that is.

I guess that means I'm a self-hating upper-middle-class guy. I must hate people (like my wife & I) who make good money, save and invest. Yup, yessiree, that's it exactly...

"This claim ignores that those same conservatives voted en bloc against pay-go shortly before that."

How did it ignore that? I would like to know so that I can be sure to do it the next time I am commenting on something completely unrelated to it.

I actually don't understand why we need tax 'brackets' at all. Why not a nice gentle but continuous curve which maxes out at 35-40% point?

Sebastian, as I said, I don't know enough about S-corps to comment on those so I should probably stop. I'm happy to drop the headline 2/3 number. My point is really regarding recycling of earnings into reinvestment, and that no taxes come into play in that event, right? Money that leaves the business is taxed, but that money isn't a factor of the expansion of the business anyway.

S-corps were designed to provide a single layer of corporate tax for federal income tax purposes to small businesses, rather the two layers of tax that are imposed on C-corporations (on the corporate income and then again when dividends are distributed), along with limited liability (before there were LLPs and LLCs). Their use is, in fact, much more complicated for tax purposes than using a C-corp.

Now, nobody sets them up anymore because everyone uses LLCs.

I think I lost my comment, my apologies if this shows up twice.

Reading over this thread, what I take away is this: the reason it's good to tax investment income at a lower rate is that investing is optional. Without the incentive of a lower tax rate, folks will do something else with their money. The economy will starve for capital, and jobs will be lost.

People who work for a living generally don't have a similar ability to keep their labor out of the economy. They need to eat.

There's a kind of "root hog or die" aspect to the labor side of this that is perhaps unfair, but if so that unfairness is more or less baked into the situation, and should probable not be accounted for in public tax policy.

That appears to be the argument.

A couple of thoughts.

First, can someone show me a solid, credible correlation between historical capital gains rates and the unemployment rate? If the argument above holds, it should be reflected in the historical record.

Next: is capital insufficiently available to the small and medium sized businesses that drive job creation? That could easily be so.

If it is so, is that because people aren't making their money available for use? That doesn't seem likely, especially at today's capital gains rates.

Perhaps there are other reasons that capital isn't making its way to small businesses.

Next: what other factors drive job growth?

One that comes right to mind is demand. It doesn't matter if it's an employer's market for labor if there isn't something productive for new hires to do.

The bottom 40% of the population pays 5% of income taxes. The bottom 60% pays 15%.

How about an absolute income tax holiday for the bottom 50% of income earners? It would cost us maybe 10% of current revenue. All or damned close to all of that money would be spent, creating consumer demand. Inventories would be cleared, new goods would need to be created, jobs would result.

But nobody is proposing that version of "forgo revenue to create jobs". The only comment on the topic are complaints that the lower quintiles are already getting too much of a free ride.

What else creates jobs?

Infrastructure. Education. Investment in R&D. All of that generally calls for public money. Spend the money, prime the pump, grow the economy.

The rising tide floats investor's boats as well as workers.

But that's socialism.

I have no resentment of wealthy people, I have no quarrel with the argument that capital is necessary to make things happen. All of that is fine with me.

But the federal tax regime in the US favors capital investors over every other stakeholder in the economy. I'm not sure that's reasonable, or fair, or necessary, or right. I'm not even sure it achieves the result folks claim for it.

The top 400 income earners in the US have an average income of $345 million, and an effective tax rate of 16%. And the citizens of Arizona can go piss in the sand, because the state of Arizona is broke.

That doesn't seem right to me.

If an S-corp is passing income to its owner (which is what S-corps do) that income is taxed at the full personal income tax rate. S-corps aren't magically escaping the tax man by funelling money to their owners.

i have an S-corp. and, there are two main ways i get money out of it. the first is standard W-2 pay. on that, i pay the full set of taxes, including both the employee and the employer's 1/2 of the FICA tax.

i also take money out as dividends. on that, i don't pay the FICA tax.

my accountant encourages me to keep the split in salary / dividends at roughly 60/40.

so, yes, there is a way to escape the tax man, but it's not a huge amount, and it's only on a portion of the money.

Seconding Ugh, the LLC is the preferred vehicle in most circumstances nowadays.

I guess the relative lack of inflation in recent years has obscured part of the original justification for lower capital gains rates. If you hold an asset over a period of moderate to high inflation, a lot of the "capital gain" you nominally get when you sell is really a return of your (inflated) principal, not actual profit. Having a lower capital gains rate helps to compensate for this difference, although not perfectly. For example, if you buy an asset for $10,000, and sell it for $22,000 after seven years of 10% inflation, you would get taxed on a $12,000 capital gain - but that $22,000 is only worth $11,289 of goods and services measured in year 0 dollars. Adjusted for inflation, you had a modest 13% profit in real dollars after 7 years, not the 120% that you pay taxes on.

I actually don't understand why we need tax 'brackets' at all. Why not a nice gentle but continuous curve which maxes out at 35-40% point?

If you plot tax due versus AGI under out current bracket structure you get something like "a nice gentle curve". Not a staircase, at any rate. Marginal rates, remember?

I have absolutely no problem with a "continuous curve". I don't see much point in it, however, since ALL the fuss and bother of doing tax returns lives in the additions and subtractions required to compute AGI in the first place. Computing the tax after that is trivial, whether you use a printed table, or a piece-wise linear function, or some scary-looking exponential function.

In the context of this thread, you're probably not concerned with simplifying tax returns so much as with reducing "distortion" of incentives -- specifically in the neighborhoods of the bracket boundaries. Two things about that:

First, even if we believe in the existence of a subspecies we might call homo economicus marginaltaxus, a subspecies defined by its sensitivity to marginal tax effects, the fact that AGI -- not "income" -- is what gets taxed means that this poor creature must spend its economic life in a total tizzy. "Is it worth earning this extra dollar of income? Not if it pushes my AGI into the next bracket. Wait: what's my AGI without this dollar?" The poor creature must be doing its tax returns real-time in its head every day of its life.

Much more important: if you plot the tax versus AGI graph over its whole range, you notice that after about $350K of AGI the curve quickly approaches a dead-straight line. That's the smoothest, gentlest "curve" you can have. No distortions there at all. The truly "productive" individuals of the species homo economicus marginaltaxus live in a flat tax world already.

Now, a truly clever member of the species might notice that there's a connection between his own tax rate and the tax rates of those farther out on the income axis. If he and they live in the same nation, and the nation needs to collect a certain total amount of tax, the lower their rate is the higher his own must be. If he nevertheless believes that people who make ten or a hundred times the income he makes ought to pay no higher a marginal rate than he does, then he's not so terribly sensitive to his own marginal rate, is he?


well carleton. I know actual flesh and blood people who make decisions on how and where to invest their money who have and would make different decisions if the capital gains tax was raised to 35%, and i know many of these people.

The plural of anecdote is not data.

Even if your assertion is true and your friends' behavior would change, the data presented suggest that their changes in behavior wouldn't affect the growth of the economy. Which isn't surprising- they weren't going to keep their money under the mattress. They'll put it into the world where they can get a return, and that means it'll be available for growing businesses.

Dave W. - it sounds like an inflation-compensated capital gains tax would be a good idea, where the each year the basis is inflated by the CPI, and the gain is measured and taxed against the inflated basis.

Wouldn't be too complex to calculate, I think.


I actually don't understand why we need tax 'brackets' at all. Why not a nice gentle but continuous curve which maxes out at 35-40% point?

Good idea. We'll give you the job of explaining to everyone that their income tax liability is just the integral from zero to their income of the tax curve.

Good idea. We'll give you the job of explaining to everyone that their income tax liability is just the integral from zero to their income of the tax curve.

No need for sarcasm. Afaik that's exactly how it works over here. And the integral can be solved leading to a calculable tax formula.
For those that do not understand the math (i.e. most public officials/employees) the results are printed in table form. This will soon end though because tax returns will have to be filed online in the future with comrade computer doing the calculations for you.

Meanwhile, since we were talking about tax disincentives to business formation, let me share my experience with unemployment. I started my own tutoring business after I lost my job last August. The problem is that unemployment reduces my benefit by 75% of whatever I gross, before I get to deduct expenses. Since my expenses have been running at around 50% of my gross (using IRS mileage rates), that means I have been losing around 25 cents on every dollar I take in, once you take the effect on unemployment into account. And that's before you take stuff like the self-employment tax and local business tax into account.

To take my most extreme example, one of my first clients was a 140-mile round trip from me. The client paid $120 for a two-hour session once a week, which was a premium rate in view of the distance. My cut of that ranged from 60-70%, which counted as my gross (the rest went to the brokerage service that found me the client and handled the billing, etc.). So, assuming the highest rate, my gross was $84. Unemployment took $63 of that, leaving me with $21, out of which I paid a $4 bridge toll and 3 gallons of gas at $3/gallon. So even just looking at the immediate cash flow, I was netting $8 for a two-hour lesson, three hours of unpaid travel, and a couple hours of unpaid prep per week. If you then account for indirect car expenses that the IRS mileage rate tries to include, like depreciation, insurance, and repairs, this was clearly a net loss financially. If I could have that $63 back, it would be a rather different story.

I still justify doing this to grow the business and provide myself with a sustainable income source (unemployment won't last forever), but it's a pretty big disincentive to the newly unemployed starting new businesses. As a society, it might be more valuable to look at fixing that problem first before worrying about the effects of small changes in the capital gains rates. It would also have the advantage of giving most of the benefit to the currently unemployed and underemployed, rather than well-off investors (which admittedly, are not completely exclusive categories).


No need for sarcasm.

No doubt it can be simplified, but I find the notion amusing nonetheless.

I have a friend who is a chemistry professor, and he told me the following tale:

Teaching freshman chemistry, he was trying to explain, without resorting to calculus, how some quantity could be calculated. Terms like "integral" tend to panic freshmen here, if not in the land of Leibniz.

He went through a spiel about ever-narrowing rectangles, and made the point well enough, as he thought. After class one of the students approached him and said, "You know, Professor, there's an easier way to calculate that. Let me show you."

Two mathematicians are eating lunch out. One, in the middle of grousing about the appalling ignorance of the general population, excuses himself and goes to the bathroom. The other calls the waitress over.

"Here, take this ten dollars. Now, when my friend comes back, I'm going to wave to you. Come over, and whatever I ask you, say 'X squared'. Have you got that? 'X squared'. Say it for me? Good, now scoot."

When the first one comes back, he says, "I think you're much too pessimistic. I'll bet you fifty dollars that even our waitress could do a simple integral. Miss, can you come here? Now, what's the integral of 2x? What's that? 'X squared?' Yes, exactly, thank you."

And as he accepts the fifty bucks, the waitress walks off, muttering under her breath "Plus a constant".

"Good idea. We'll give you the job of explaining to everyone that their income tax liability is just the integral from zero to their income of the tax curve."

You're overcomplicating. 99% of people will just look up their owed tax on a chart.

You're overcomplicating. 99% of people will just look up their owed tax on a chart.

Yes, I am and they will. (Though if the curve is really continuous the chart will have to be quite large, I think.)

It would still be fun to hear someone explain it, maybe in Congress while they're debating the bill.

You're overcomplicating. 99% of people will just look up their owed tax on a chart.

Ah, there's the rub. The tax tables are currently how most people compute taxes, so why on Earth would they need a tax function whose first derivative is continuous, when a piecewise-continuous function serves as well?

"Brackets" as a descriptor for the tax-rate breakpoints is a misnomer, I think.

Like Bernard, I think it would be very amusing indeed to hear the tax-rate debate shifted to coefficients and polynomial orders. Imagine bills of attainder represented by the addition of a 1/(Income we don't like) term in the tax equation, or similar.

Should I march out my inverse tangent proposal again? The shape of that curve just seems so right to me. Plus it would give people an incentive to learn trig! (Like they should need one. What's more fun than trigonometry? I mean, seriously! Right?)

What's more fun than trigonometry? I mean, seriously! Right?

Complex variables. It's the font from which all trig identities flow.

Oh man, I remember one of my last math exams at school where the old German income tax formula was the topic. The equation for each bracket became a variable in the equation for next higher bracket and all of it was in written out text. The result, when written using normal mathematical notation was quite a complicated rational function. I think some tax laws would also need a clarification that of all solutions the equations have, only the positive and real ones are to be considered. "You owe us (3482 + 925i)€ would make things slightly complex ;-)

"You owe us (3482 + 925i)€ would make things slightly complex ;-)

Sometimes it seems like some of my money is imaginary.

"You owe us (3482 + 925i)€ would make things slightly complex ;-)

Oh, I would totally write a check out for that amount.

Oh, I would totally write a check out for that amount.

I'll give you e^(i*pi)+1 dollars if you do.

Complex variables. It's the font from which all trig identities flow.

True, but that's like skipping plane geometry and going straight to analytic geometry. So many of the really beautiful proofs are replaced by dead simple calculations.

I've had too many other people give me that much, hsh.

The comments to this entry are closed.