Why did American taxation become less progressive?

That is an underexplored question, and it is considered in the job market paper of Chunzan Wu, from the University of Pennsylvania.  Here is the abstract, which will unsettle many people:

Since the 1970s, income inequality in the U.S. has increased sharply. During the same time span, the U.S. federal income tax has become less progressive. Why? I examine this question in a Ramsey optimal tax policy framework. Within this framework, the tax policy is determined by: (1) a set of Pareto weights representing the government’s preference over different households; and (2) household lifetime utilities summarizing the effects of economic fundamentals. I first study the changes in economic fundamentals using an overlapping generations incomplete-markets life-cycle model with heterogeneous households. The model features both endogenous human capital accumulation and household labor supply and is calibrated to the U.S. economy in the 1970s and 2010s. Then I use this economic model to determine whether the change in income tax is the result of an optimal policy response to changing economic fundamentals or the consequence of a change in Pareto weights. I interpret the latter as changes in the political influences of various income groups. I find that: (1) changes in economic fundamentals alone induce a less progressive optimal income tax and can account for 40% of the reduction in progressivity we observe; and (2) the change in Pareto weights required to explain the remaining part of tax policy change favors high-income households and also implies less valued government services. Finally, using a stylized political economy model, I discuss potential explanations for this change in Pareto weights such as the lower cost of conveying information to swing voters and the rising inequality of voter turnout among different socioeconomic groups.

The paper is here (pdf).


Krugman, et al.: Evil Republicans!

Well let me suggest the alternative explanation: Evil Democrats.

The Democrats' voter base is dominated by the poorest people in America and the richest. North-Eastern liberals and poor Blacks. This seems a difficult situation to maintain. The solution is obvious - talk about higher taxes but grant preferential tax breaks to your favored clients. That way you can tell the poor you are voting for higher taxes while telling the rich that they don't have to worry because they won't pay any.

Does anyone think Hollywood, for example, pays close to the average rate of tax?

"The Democrats’ voter base is dominated by the poorest people in America and the richest."

Not true, read Gelman.

Could you tell us what he says instead of directing us to read a prolific person's entire body of work?


Rich people vote republican, poor people vote democrat ... but the slope of the line depends on how rich the state is.

In richer states, the slope is flatter, so rich are more likely to vote republican but not by much.

In poorer states, the slope is higher, so rich people there are much more likely to vote republican.

I love Gelman's work but I am not sure it directly refutes the above point. From memory Gelman divided each state in 3 tiers and looked at the relative in state voting patterns of top-middle-bottom. His results don't really preclude the hypothetical that a significant number of the very rich are voting Democrat and that policy decisions resulting from this. Correct me if I am wrong.

Hollywood is a drop in the bucket compared to the finance industry. Most favors to Hollywood come in the form of IP protection, not marginal tax rates.

In 1979, the richest fifth made 45 percent of income and paid 55 percent of taxes. Today the numbers are 50 percent and 70. That is all federal taxes, not just income taxes. American taxes have become more progressive, not less.

You are reporting what portion of taxes came from the upper quintile, not the relative percentage of the incomes of various quintiles went to taxes. The latter is how one measures progressive taxation. The former can be explained by, for example, greater income inequality. As a greater share of a nation's income goes to the wealthy, a greater share of the nation's taxes will come from the wealthy, all else being equal.

So, its not about revenue, but punishing the right people. Got it.

But I think the paper accounts for this fact, in the "40%" figure. Also the richest fifth make 70% of income today as well as pay 70% of taxes (I do recall the figures are symmetrical, not sure about your 70% but it's symmetrical).

Seems the rich and the middle class both got sick of government in the 1970s, and turned to the Republicans. Unfortunately due to government spending the tax cuts were not sustainable. But, as the paper implies, the voters did not care. Hence AARP voters are likely never to cut Social Security, even if it's not sustainable.

Not sure about the 40% of the paper--is it saying the hollowing of the middle class accounts for that factor?

Cassander is leaving out state and local taxes, which are regressive.

As of 2011, the top 20% earned 59.6% of the income and paid 63.1% of all taxes at all levels, not too far apart. The top 1% earned 21.0% of the income and paid 21.6% of all the taxes, pretty much dead even.

Of course, the top marginal fed income tax rate from 1979 to today has declined from 70% to under 40%, and between 1940 and 1964, the top federal marginal income tax rates were usually over 90%. Sorry, Cassander, but the US tax system has become substantially less progressive.

I disagree.

The value of the top bracket rate is a terrible metric of progressivity, because it ignores deductions the bracket boundary and most importantly because it ignores the other brackets. The main idea of progressivity is how the effective rate changes between the poor and the rich, so any reasonable metric is going to need to compare those numbers. A very simple metric is the difference between the effective rate between the lowest and highest income quintiles, and under that metric the US federal income tax has become substantially more progressive (http://taxfoundation.org/article/cbo-report-shows-increasing-redistribution-tax-code-despite-no-long-term-trend-income-inequality).

If you disagree, please define and defend for your preferred metric of progressivity and cite data showing that value is lower in 201x than in 197x.

According to the CBO, in 1979 the top 1% paid an average total federal rate of 35.1%, the middle three quintiles paid an average rate of 19.1%, and the bottom quintile paid 7.5%. In 2013, the top 1% paid an average rate of 33.3%, the middle three quintiles paid 13.3%, and the bottom quintile paid 2.9%. That looks more progressive to me.


Great stats.

It also shows you where the taxes will have to rise to pay for all the goodies people claim they want (we all want lobster and steak until the price is shown to us.)

Like Europe, that means swingeing middle class taxes like VATs.

Your data at least includes fica but excludes state and local. My numbers cover all of them. Sorry, while you are better than David Wright, you are still not on top of what is really going on.

1. The paper in question made a specific claim about federal income taxes.
2. You gave no link to your numbers. I'd like to see the methodology.


Google Citizens for Tax Justice. They have the numbers.

If this is what you are referencing, their methodology seems bizarre to me. They're taking a broad measure of tax incidence and combining it with a narrow measure of "cash income," which seems like an obvious attempt to inflate tax rates paid at the lower end. On the other hand, they include "corporate profits net of taxable dividends" as part of a stockholder's personal income when calculating his or her tax rates, which seems like a strange attempt to create the opposite effect.


The CBO, Joint Committee on Taxation, and the left of center Tax Policy Center all have top earners paying total federal rates in the low to mid 30's right now. Citizen's for Tax Justice has them paying an average federal rate of just 24.3%. CTJ also has the lowest income levels paying higher rates than all of those other sources do. In short, CTJ is an aggressively partisan organization whose methodology appears to be well outside of the mainstream.

In any case, I don't see any long-term historical numbers from CTJ, so it still doesn't answer the question of whether rates have become more or less progressive since the 1970's. In the case of the federal income tax (which is what the original claim in this post was about), it seems clear to me that it is in fact more progressive now.

The most progressive income tax system consists of taxing Bill Gates and Warren Buffet a dollar each, and everyone else zero.

Your numbers are simply about federal income taxes. They do not count regressive fica taxes, which is the largest tax paid by about 3/4 of the population, not to mention also regressive state and local taxes. Why is it that so many people going on in your vein make this same silly mistake?

1. I was countering an assertion made in the paper specifically about federal income taxes, so it seemed reasonable if not incumbent on me to limit my facts to federal income taxes.

2. Even if one does include all federal taxes, my assertion is still true. (http://www.taxpolicycenter.org/taxfacts/displayafact.cfm?Docid=456) Bottom quintile all-up effective federal tax rate went from 7.5% in 1979 to 1.9% in 2011; top quintile all-up effective federal tax rate went from 27.1% to 23.4% over the same period. The difference in rates between top and bottom quintiles (my simple progressivity metric) went up from 19.6% to 21.5%. FICA, by the way, really hasn't changed that much; it went from 5% to 7% for the bottom quintile and stayed about 6% for the top quintile.

3. State and local taxes are all over the place (http://taxfoundation.org/blog/which-states-have-most-progressive-income-taxes-0). I don't know of any study which has tried to aggregate this data across the nation while perserving income quintile distincions; do you? In any case the relevant information here is not "how regressive on average are state and local taxes right now?" but rather "how has the average regressivity of state and local taxes changed over the last 40 years?" Can you even claim to have a reasonably confident guess about the sign, much less the magnitude, of that change, averaged over every locality in the country, over the last 40 years?

Even definitions like "high tax state" or "low tax state" shift dramatically over time.

Massachusetts used to be derided as "Taxachusetts" but decades of Republican governors and a constitutional commitment to flat income taxes have kept Massachusetts off the top ten list of highest taxed states.

Connecticut, meanwhile, had no state income tax as late as 1991! Now it's one of the highest taxed states in the nation (combining property, sales, income, etc.)

Perhaps, but the original paper mentions the federal tax only. Still seems to relevant if you're saying that the original paper was wrong. It does, for example, in Appendix A appear to claim that the Bush tax cuts made the federal tax more regressive, but that's under a very strange definition of regressive. On the whole. the Bush federal income tax cuts made the federal tax more progressive (slightly) just lower, as the graph itself shows.

Then again, perhaps the paper just doesn't define what "progressive" means, and considers any tax system to be more "progressive" if the top marginal rate is higher, even a system (as in Europe) where the marginal rate ascends steeply and the top marginal rate is reached on a far lower income than in the US. European progressivity is on the spending side; on the tax side Europe taxes the middle class much higher relative to the rich.

The cherry picking is always so careful when talking about such underexplored questions.

How did the American capital gains tax rate develop between 1980 and 2015? Hint - a third less, actually.

And how do wealthy people prefer to get compensated? It isn't income, at least in terms covered by income tax.

An example of how that works in practice - 'The Romneys -- like many taxpayers -- received a tax refund for their 2010 filing. But at $1.6 million, their refund was slightly higher than the IRS-reported average of $3,003.

In 2010, Romney and his wife totaled $21.6 million in income and paid slightly more than $3 million in federal income taxes. That's good for an effective tax rate of 13.9 percent, according to the documents he released. Last year, according to estimates Romney released, the pair took in $20.9 million, on which they will pay $3.2 million -- 15.4 percent of their income.

(The Obamas, by way of comparison, had $1.7 million in income in 2010 and paid an effective tax rate of 26.3 percent.)

So how does a man worth hundreds of millions end up with a tax rate similar to that paid by a household earning $50,000 per year?

The answer lies not in any shady accounting or financial trickery, but in the federal income tax code, explained Joseph Newpol, a professor of law, taxation and financial planning at Bentley University in Waltham, Mass.

“Essentially, the policy is this: if you have income from capital, we’re going to tax it at a preferential rate,” Newpol said.

Most workers are familiar with the progressive income tax: Money earned in wages or salary is taxed at a higher rate the more you earn. For 2011, a married couple earning a total of $50,000 – after deductions – is subject to a maximum tax rate of 15 percent; those earning more than $379,150 will pay 35 percent on some of their income.

But capital gains – the profit made from selling an asset, such as real estate or stocks – are taxed differently. Though there are some exceptions, this money is generally assessed at a rate of 15 percent.

For tax purposes, other income also falls under the capital gains rate, including carried interest, which is when an individual is paid a portion of the profits generated by a fund or partnership.

In Romney’s case, $12.6 million of his 2010 income was classified as capital gains, according to his tax filings. Had that money been paid as salary or wages, most of it would have been taxed at 35 percent; instead, it was assessed at less than half that rate.' http://www.salary.com/why-mitt-romney-other-wealthy-investors-pay-less-taxes/

There is no question that the American tax system has become less progressive - but using income tax as a way to measure it actually misses a significant aspect of what the American economy has become. A shell game, with enthusiastic cheerleaders getting crumbs for their efforts to ensure the rich get richer.

Welcome to those small steps towards a much better world.

Your analysis ignores the Obamacare surtax that went into effect in 2013. Also the very significant charitable donations that the Romneys give.

That should be "charitable". Giving money to the Mormon church so he can be sure he'll get his own planet isn't exactly feeding the hungry.

Nice touch of bigotry Ed

I'm sorry this isn't a safe space for you V D, maybe you can go find a college administrator to hear your pain.

Sorry, VD is right, the Mormon Church is as good a charity as any.

So you must be even more angry about the existence of the tax exempt government bonds, then, which allow people to avoid paying tax at all on income? That is, after all, a large part of the reason why the AMT was created, a few wealthy people who invested primarily in government bonds avoiding all taxes.

Or perhaps you understand that in that case the yield on the bonds are lower to compensate, and the real subsidy ends up being towards government, so you support them. Perhaps a similar level of sophistication would be appropriate when looking at capital taxation. For one thing, the lower tax rate is on long term gains, which are not otherwise adjusted for inflation.

For taxes in general, it seems strange to praise high marginal tax rates and ignore the effect of decreasing deductions and exclusions-- including some very large above the line exclusions that were reduced in the 1986 Tax Reform, causing reported gross income to be increased.

There are other changes as well. For example, company cars are still very common in the EU but have become much less common in the US, thanks in large part to marginal income tax rates. Those don't show up in income, but I think we can agree that someone with a nice car as a perk is better off than someone without.


The Left claims they want rich people to pay taxes, especially the idle rich, but they always leave the tax-exempt bonds alone. I do not understand that.

The tax exemption of government bonds is largely a wash. If we taxed that interest, the government would have to offer higher interest rates to attract investors.

So would we rather have lower tax revenue and lower bond interest payment or higher tax revenue and higher bond interest payments?

Clearly we'd want higher tax revenue and higher bond interest payments.

That would put the cost of inflated spending on the states doing the spending, not spread out across victims in other states.

While we're on the topic of terrible tax policy, we should also drop the federal deduction for state and local taxes. It's the worst kind of distortion, encouraging exactly the wrong thing at lower levels of government. It's as if "burglary tools" were made deductible - it makes no sense.

Income seems like it would be simple to define but it is not. Tax incidence is tough. Who consumes less due to Corporate taxes, Matching FICA property taxes is not so simple to discern. If you own a corporation that paid Corporate taxes (and matching FICA if you naively believe we who writes the check pays the tax) and then Capital gains taxes should both be included in calculating the taxes you paid?
That is why I think a progressive consumption tax would be best. I agree with Scott Sumner that progressive consumption tax could have a top rate of 70% and still not hurt the economy much.

Everyone knows that tax rates of 50% or more discussion discourage hiring workers or paying workers to build capital because the tax deduction for this spending is only 50% or more.

The way to create more jobs is to eliminate income taxes so there are no tax dodges for hiring or building capital assets and wages are paid entirely out of profits.

I know this is true from listening to John Boehner who based on his experience in business knows based on the explanations of his tax accountant that income taxes must be paid on wages paid and and other labor costs.

I also know I wish I had been his tax accountant getting rich pocketing the money I was telling him was going to the IRS.

Ps, I remember Milton Friedman's attack on high tax rates and related on PUCs for the same basic reason - both encouraged wasteful hiring of workers that lowered productivity and drove wages up too high, dragging down the economy.

With high tax rates, paying added employees cost too little because the Federal and State IRS paI'd 50-60-70% of their wages. And electric and telephone utilities hire too many workers and build too much capital assets so the utilities charge too much providing too much service that is far too reliable, because most people would prefer waiting to be connected or to have service go out and stay out for days and weeks after storms to pay lower rates. Just like taxpayers want the roads to have potholes and to be excessively congested with dangerous bridges that might fall in order to have lower taxes. Basically, PUC central planning rate setting based on labor costs (capital is time shifted labor costs) leads to excessive payment to labor to increase profits, so it's just like high tax rates which promote excessive payment to labor.

The basic "stylized fact" asserted by this paper is complete bunk: the federal income tax has become noticibly more progressive over the last 30-40 years. You can find this data just about anywhere, but here's one source I found via a quick Google search:


Basically, the effective rate for the top income quintile has stayed at ~15%, while the effective rate for lower quintiles has dropped, most precipitously for the lowest quintile for which it went from ~0% to ~-10% (mostly, BTW, due to the expansion of the EITC).

I have no idea what metric of progressivity the article is using. Given that the "fact" it purports to explain is the opposite of what every tax wonk knows, I expected to find a detailed discussion explaining and justifying its unusual metric, but I can find no such section. Can someone who has combed through this work more carefully enlighten us?

Sorry, guess I am going to repeat mysefl, but if you are going to repeat yourself with bunk, it needs to be pointed out again. Your data is only for federal income taxes. this is completely misleading. You should be ashamed of yourself, but I have no doubt that you are not.

David: "the federal income tax has become..."
Barkley: "Your data is only for federal income taxes. this is completely misleading. You should be ashamed of yourself..."
David: !?

Could you please show your work (with references) including all taxes? If the issue is whether all taxes have become less progressive since the 1970's, then I guess it would be necessary to show not only that the effective tax rate for the top income percentile has declined, but that it has declined more than other income percentiles. It would also be necessary in this regard, I think, to use a consistent definition of "income". By "income" do you mean economic income or only the income that the tax code, rather arbitrarily in many cases, classifies as "taxable"?

Next step, I think, would be to compare this with the effective spending. If one asserts that FICA and Medicare taxes have become more regressive over time, would it also be relevant how these targeted "taxes" are spent". In this regard, is it at all important, in your mind, whether the "contributions" for Medicare have become more or less "progressive". Should the "taxes" paid for Medicare post-retirement be considered "taxes" any less than pre-retirement contributions?

Finally, you earlier comment seems to suggest that in your mind the real problem of decreasing progressivity since the 1970"s is between the top 20 percent and the top 1 percent. Is that so?

If you don't answer, I would be disappointed, but there's no need to "feel ashamed of yourself".

Here's an estimate from the Tax Policy Center that includes federal income, corporate and payroll taxes (but excludes excise, estate, tariffs and state and local):


The estimate includes the effects of the last tax hikes for 2013 (why stop at 2011??)

Yeah, only does quintiles, which misses the super regressivity at the top end. My data is from Citizens for Tax Justice, although they got it from official data sources. Sorry, 2011 all that is available. Not much change in 2013 though.

Gee, Barkley, now I'm really disappointed. Your answer is non-responsive and the link I referred you to clearly includes more than just quintiles. See under "addendum" immediately below the quintile information.

It isn't only quintiles, Barkley. It goes all the way to the top 0.1%.

Where's Barkley?

He seems to have mysteriously vanished from our little conversation above too. I'm starting to notice a pattern with him.

The pattern that caused me to jump in is that he is fond of telling people how "ashamed of themselves" they must be for disagreeing with him, not only on debatable issues, but even when he's clearly wrong.

I don't understand why the marginal income tax rate increases steadily up into the $400ks and then stops going up. If a person making $500k can afford to pay a higher marginal rate than somebody making 50k, then why not somebody making $5 million or $50 million paying a higher rate than somebody making $500k?

Under models for progressive income taxation based on diminishing marginal utility, this makes sense. But in the real world, a few more factors come into play. The current top bracket only hits ~1% of the population and thus changes to hit make only small changes in income tax revenue; your new-top bracket might hit only ~0.1% and thus make even less difference in revenue. That's an awfully big political fight to have for an awfully small revenue gain. Also, at some rate you hit the peak of the Laffer curve for these individuals. There is a lot of disagreement about exactly where this peak is, but since the current top rate is already ~40% and your new rate would presumably be substantially higher, you need to at least start worrying that your new rate might be close to that peak and maybe even on the wrong side of it.

" changes to hit make only small changes in income tax revenue"

The argument isn't about raising funds to finance government, it's about punishing those that have more than others, since the federal government can enpixelate all the money it needs to build an aircraft carrier or pay the salaries of an army of state department drones.

It's very rare to earn $500,000 on wages and salary. If you are pulling that down its likely capital gains, or sweat labor you can convert to capital gains (like a doctor with a solo practice).

Rare, but not impossible. Probably the highest wage earners are athletes. They could, but perhaps because of high time preference don't, make deferred compensation agreements but it is hard to see how they could convert much income to capital gains. In any event there are plenty of wage earners making $100,000 which is sufficient to have a higher marginal tax rate than someone whose entire income is derived from qualified dividends and long term capital gains.

$500k is only $250k per person in a two-earner household. Most doctor couples probably make that.

True, but many doctors work in some sort of arrangement that isn't pure W-2. Also moving to a couple seems a bit like fighting the hypo.

I was just thinking about it from a tax rate perspective. Above middle-class rates you're back to single and married-filing-jointly having the same marginal rates at the same income.

It isn't that rare on the coasts. Couples in both medicine and in tech can hit this with more frequency than one would think.

Further, in San Fran and in New York a considerable portion of single earners hit this when considering Total Cash Compensation.

Like I said, a doctor with a practice can convert part of his income stream to stay inside his business for years.

39.6% isn't the the top of the Laffer curve (yes I know its higher state by state). Law partners, finance folks, and NFL stars aren't going to curtail their behavior if that rate jumps up a few points. We'll just get more revenue.

"39.6% isn’t the the top of the Laffer curve (yes I know its higher state by state)."

Which makes it over 50% for California, NY, etc. And there's been research that the Laffer Curve slope is flat by 70%. It starts tapering off before then.

"that rate jumps up a few points."

Sure, but how many people who are actively touting a tax hike on the top bracket want a jump of just a few points?

Are you aware that at the top end the total tax system of the US is highly regressive when looking at all taxes? Those who pay the highest average rates are at about the 96th to 98th percentile of income. Above that the average rate declines, and it keeps on doing so as one's income rises, so that when we get to that tippy top sliver of the top 10 or 1% of the top 1% their rates are much lower than those further down in the income scale. This is the basis of those stories about Warren Buffett (accurately probably) noting that his secretary was paying more in average tax rates than he was.

It is the cap on fica and the regressivity of state sales taxes that are the main reason for this. But I guess you are unaware of this fact. Deal with it before you keep on with this silliness.

I don't think it is sales tax or FICA that is the driving factor here. Then the peak in rates would probably come in the low 90th percentile. What I think is actually going on is more and more income is not W-2 and so it is taxed at different rates as you get higher and higher up the curve.

The assertion that "at the top end the total tax system of the US is highly regressive when looking at all taxes" is easy to dismiss even if I accept your requirement that I look at the top 0.01% and include all federal, state, and local taxes. I can say this with high confidence even though I don't have aggregated state and local data. Here's how: The way to pay a low federal income tax is to have it all classified as capital gains, which is currently taxed at 20% for the top bracket. Suppose the entire top 0.01% manages to do this, and that they pay no FICA other federal taxes. Their federal effective tax rate is then 20%; this is pretty much the lower bound for a very high income federal tax rate. We already know that the lowest quintile all-up federal effective tax rate (including FICA, income, et al.) is 2% (http://www.taxpolicycenter.org/taxfacts/displayafact.cfm?Docid=456), so to make the "total tax system" regressive, state and local taxes are going to have to be so regressive so as to erase this 18% gap. Total state and local taxes are about half as much as total federal taxes, so, very approximately, average state a local taxes would need to be about 36% on poor people and 0% on rich people to make the "total tax system" flat (much less "highly regressive"). Can you name even a single state with a documented tax burden either (a) this regressive or (b) this high at all, for anyone? Do you define "highly regressive" just as "not as progressive as I want"?

Here is the 2015 data. Overall system just slightly progressive. This link does not show what happens within the top 1%, but I note that fica rate simply steadily falls as does state sales tax rate, and nothing else is rising as a rate. Oh, and long term capital gains tax rate is only 15%, David. You did not know that? Sheesh.

ctj.org/ctjreports/2015/04/who_pays_taxes_in_america_in_2015.php#.VI_7Eb9OL1U .

Sorry if that does not work, but it can be found by just googling "Citizens for Tax Justice Who Pays Taxes in America."

Sheesh, Barkley. The long-term capital gains rate has been 20% for the top bracket since 2013. (https://en.wikipedia.org/wiki/Capital_gains_tax_in_the_United_States#Regular_and_capital_gains_tax_rates_for_2015.5B5.5D)

Sheesh, David, I guess you nailed me on that one. However, nobody in the top group has all their income stated as cap gains, so declaring that the cap gains rate provides a "minimum" is inaccurate. There are a variety of other ways very high income people are able to get their average rate still lower, and indeed this was the basis of Warren Buffett noting he had a lower rate than his secretary. I may be mistaken (feel free to rip me into little shreds if I am), but my memory is that in fact both Buffett and Mitt Romney had average tax rates lower than 20%.

Well, sheesh. You are both wrong.

If we are talking only about federal taxation, the current marginal tax rate on capital gains is 23.8 percent. Both of you forgot the 3.8 percent ACA Medicare surcharge.

I believe Buffett is using taxable income rather than total income as the base when he compares his and his secretary's alleged tax rates (he does the same thing with the so-called Buffett Rule proposal). He also includes the employer's side of FICA as part of the employee's tax rate, but completely excludes the corporate income tax from his own.

It would be great if they'd at least pay the *same* rate instead of a significantly lower one.

Would it? At the high-end, a lot of that income is capital gains where we want the rate to be zero, and certainly not the same as the income rate.

We want it to be true of true capital gains. But does everyone who pays the capital gains rate truly invest the money to start with. Should hedge fund manager pay the capital gains rate or pay the income tax rate?

(The question is serious. I have no idea how the taxes are structured in those cases. But it seems like capital gains should be reserved for money that's been taxed previously.)

I'd agree that there are some confusing issues at the margin, but I don't think that's what brad was talking about.

It's a decent argument for tearing down the whole edifice and replacing it with a VAT. Replacing, not supplementing.

Begs the question and/or equivocates on 'we'. I don't want the capital gains rate to be zero, I want it to be the same rate as earned income after adjusting for inflation.

Who is "we," "Lord Acton"? Gag.

Action, bub.

Sorry if it wasn't clear. By "we," I meant economists, and more generally people who care about the fairness and efficacy of tax policy.

I realize I should have mentioned there are people out there who just want to tear down the successful, and they don't like the idea.


Yeah, immoral meanies like the late Jesus Christ , who declared that, "It is harder for a rich man to enter heaven than it is for a camel to pass through the eye of a needle."

If it's double taxation you object to you should prefer to get rid of the corporate income tax, which faces intractable measurement and collection issues, rather than eliminating the taxes on capital gains, dividends, non-cash compensation and so on to flesh and blood human beings.

This study is limited to federal income tax, while nearly 80% of working Americans pay more in payroll tax than income tax. As the federal income tax has been cut, the federal payroll tax has been increased (and increased a whole lot by the 1980s "reforms" that were phased in over many years). Politicians sell "cuts" in taxes that are aimed at (high income) payers of the income tax, while most votes don't appreciate the difference: to them, "taxes" are the amounts withheld from their paychecks. This combination, promises of "tax" cuts while experiencing "tax" increases, had the effect of making tax protesters out of moderate income voters (hence, the real genius of President Reagan). Moreover, middle income taxpayers were hit the hardest by the payroll tax increase as it raised not only the tax rate but also the tax base; indeed, the combination of the payroll tax and the income tax results in working Americans earning around $100,000 to $120,000 paying the highest marginal tax rate. [Of course, the enormous increase in the payroll tax adopted in the 1980s generated almost $3 trillion (that's trillion) in social security surplus, a surplus that doesn't exist, it having been spent on everything from farm subsidies to wars, adding further insult to moderate income Americans who take out their anger at, you guessed it, Democrats. Again, it's the genius of President Reagan! Or was it the genius of Senator Dole, who sold the payroll tax increase to Congress in the 1980s and who now suffers the ire of, ironically, the Republican tax base (for denying Saint Ronnie the White House in 1976)]. ]

Don't worry, the payroll tax is now currently running a deficit, and has for the last five years, as you can see in its annual report. The "surplus" credit built over the years has not been fully redeemed yet, but it is going in the other direction is going now. Presumably this should make you happy.

You could cut the payroll tax, if perhaps Democrats were actually willing to cut the Social Security benefits of those with high lifetime earnings. However, reducing the benefits of the rich while protecting the poor is mostly a conservative idea. The Democratic preferred alternative is to increase payroll tax on the high end.

"However, reducing the benefits of the rich while protecting the poor is mostly a conservative idea."

Yet once some of these same people get their way, they start railing against the culture of dependency, the "takers", the 47%, etc.

The article you linked to mentions this but benefits are already substantially scaled back for those of higher incomes. Social Security is designed to be progressive and it is. You could cut further and make it even more progressive, of course, but at some point, you are either going to increase misery in old age or else kick more people onto other programs like Medicaid because their incomes no longer stretch far enough to cover nursing home care or health expenses. You can't cut $1,000 from Social Security and expect there will be $1,000 in savings in overall government spending.

You do realize, after all, that it is impossible to pay for the wishlist of the Democratic Party and its candidates (including maintaining all Social Security payouts, including to high lifetime earners) without significantly higher middle class taxes in some way? Including on those "working Americans earning around $100,000 to $120,000" who, after all, are in the top 10% of income and upper middle class to "rich."

That is why when push comes to shove, Democrats in Congress allow themselves to be "sold" on middle class and upper middle class tax increases time and again, because it's simply necessary to pay for everything. Just look at Europe, people who make $100,000 to $120,000 certainly pay higher tax rates than the US.

"Just look at Europe, people who make $100,000 to $120,000 certainly pay higher tax rates than the US." If you assume that a Briton on GBP120,000 is on roughly the same income, adjusted for PPP, as your American on $120,000, then you might care to know that the Briton pays 60% income tax on his last GBP20,000 of income. Oddly, his marginal rate thereafter falls to 40% before rising again to 45%.

Why would you use PPP?

A lot of that is because of the 20% VAT.

"Why would you use PPP?" Because it would be foolish not to?

Re: Europe, is that true?

If you need a Democratic source to say so, Austan Goolsbee says so. http://goolsbee.blogspot.com/2015/11/the-size-of-government-and-fate-of.html

From the linked article about Bernie Sanders spending proposals:

"To put that in perspective, the IMF numbers say that in 2014, U.S. government spending was around 36% of GDP . We have always been lower than the big social welfare economies of Europe. An increase in spending of the amount Sanders proposes would put us at 47.5%. That would actually put us, comfortably, in the range of the European countries, i.e., something quite different than anything in the U.S. historical experience."

The biggest difference is Europeans who make around $50K. This is median income in the U.S., and people below the median in the U.S. pay almost no income taxes, but much higher than the median in Western Europe, where median taxpayers also shoulder a heavier burden.

Thanks for replying, rayward. You're right on all counts in your reply.

Social Security keeps going bankrupt, which results in tax hikes.

Interesting to see this as a function of overall government taxation (and spending) at the federal, state, and local levels. My sense is that you will get a more regressive system as overall taxation increases.The wealthy already have the governmental services they need. The marginal utility of additional taxation and spending is much lower to them than it is to a poor person. Therefore, if the [collective federal/state/local] government wants to raise taxes, the wealthy are going to make sure that the cost of those raises is not just being borne by them (i.e. that the rises are more regressive). At the individual level, it isn't progressive/regressive that people primarily care about-it's the overall amount they are paying.

I blame mulp

My litmus test is simple. Do you want the rich and wealthy to pay more in taxes and eliminate the largest tax dodge ever?

Get rid of tax-free government bonds.

If you want to keep those, and yet rail against "the rich" not paying their fair share of taxes, you're not serious.

Given the interest rates tend to be lower on tax-exempt bonds than they would be otherwise, most of the economic benefit falls to the debtor states and not to the holders.

Largest tax dodge ever? I doubt it. The JCT estimates the cost of the exclusion for tax exempt state and local bonds to be about $23 billion in 2014. Compare that to about $69 billion for the EITC and roughly $100 billion for deduction of home mortgage interest and taxes. And, as Rob42 points out, roughly half of that $23 billion is actually a subsidy to state and local governments, not bond holders. Not for nothing does the JCT put this one under "General Purpose Fiscal Assistance". I agree it should be eliminated, though, in part because it is a "spending dodge".

For some perspective, you should review the list here:


Step 1) Subsidize government loans by making them tax-free

Step 2) Complain that people subsidizing government by accepting the lower returns on tax-free bonds are dodging taxes

I define income tax progressivity as a growing share of tax relative to share of income. The end result is that the tax system has become more progressive even before the latest hikes in the top bracket. I demonstrate here:


When you consider that a) the '86 tax reform revealed income that had been hidden (you could depreciate real estate and report income net in AGI) and the growing share of income that moved off separate corporate returns and into individual returns (Sub-S,LLC), the effect would be larger than shown above.

The only place the tax system became less progressive was within the top one percent.

Just to remind everybody here, even though some here have claimed that what the top marginal tax rate is has nothing to do with the general pattern of regressivity or progressivity (somebody here actually made such a silly point), between 1940 and 1964 the top marginal federal income tax rate was usually over 90%. Now it is true that there were a lot of loopholes so many in that bracket did not pay that rate (and I am not at all supporting going anywhere near back to such a system), the US economy grew more rapidly in that period than it has since the top maginal rate was pushed well under 50% in 1986 (where it has stayed ever since, despite some fluctuating up and down a bit). No, I am not saying that those high rates were responsible for that better performance. But I do note that they did not prevent it from happening, whereas much current rhetoric suggests that any increase in tax rate for top earners would somehow lead to horredous outcomes, sort of as many said that the tax rate increases early in Clinton's presidency would lead to a recession, and instead we had the highest GDP growth rates of the US economy since the mid-80s.

If you count the loss of benefits and tax credits as you move from low income to middle income, the U.S. tax system is brutally progressive with implied rates well over 100% in that income range. This contributes to inequality because anyone who is discouraged from trying to become middle class adds to the inequality statistics.

"Inequality of voter turnout"?! You people are out of your mind.

I'm not sure how they define "progressive taxation" but the top earners pay a larger share of taxes than ever before in history.

Of course it goes without saying that if one's complaint is that they also earned a greater share of income then one also believes such income disparities are a moral concern, but this point of view always founders on the rocks of value production: producing more income in a free market means producing more consumer surplus and raising everyone else's living standards, not reducing them, so to the extent a moral dilemma is posed by "income inequality" it is that some people are not producing as much value as they "should" (a notion that should be rejected almost as soundly as the notion as the notion that value production could be immoral).

Comments for this post are closed