Reducing inequality: what’s the problem?

“Fairness”?

Th top 1% of income earners pay 39% of all income taxes.

The top 25% pay 86%.

The top 50% pay 97%.

What could be “fairer”?

**

What should we do?

Cut the tax rates on the top earners/producers and, as history proves, revenue increases to the treasury to pay for social programs (charity).

DonC: “Cut the tax rates on the top earners/producers and, as history proves, revenue increases to the treasury to pay for social programs (charity).” History does not prove this, it proves the opposite. The Reagan tax cuts and the Bush tax cuts resulted in massive deficits, and both coincided with cuts to social spending (with the exception of the Bush prescription drug benefit, which still came amidst cuts to all other social spending).

Lane:

Regarding fairness, I’d also point out that inequality is not being spread across even the elite evenly – the wealth for the last thirty years has increasingly been flowing to those engaged in the least useful endeavors, such as the FIRE (Finance, Insurance, and Real Estate) sectors, rather than even to those elites who are involved in actually creating things of value to society. So there’s a deformation even at the top.

In regards to the impact of inequality, two obvious things spring to mind that aren’t in your list. The first is political – classical small-r republican theory teaches us that massive inequalities of wealth are bad, for reasons that go beyond “political participation, political influence, and political polarization.” Classical republicans believed – and I would argue, rightly – that economic power and political power bleed into each other, and that the existence of great concentrations of wealth damages the presumption of equality that a republic is based on. Even leaving aside the issue of direct influence, I think it’s done damage to our political imagination – he rich have been given huge amounts of deference in policymaking because we’ve valorized the corporation as the most efficient institution and the businessman as the source of innovation, we try to turn governments into businesses because businesses are deemed the best form of organization, and we seek for government-through-CEOs with the election of people like Mitt Romney, Mike Bloomberg, Arnld Swartzenegger, etc. who run on a platform of making government run like a business.

The second is economic, and just re-emphasizes the point that the marginal propensity to consume is important here. Rich people don’t spend enough to create a solid foundation of stable consumers – hence, we’ve had to turn to other sources, notably a massive binge on credit to try to make up the difference. Well, look at the result now.

“The Reagan tax cuts and the Bush tax cuts resulted in massive deficits, and both coincided with cuts to social spending.”

That is false.

What hard evidence do you have that shows social spending decreased under either Reagan or Bush the elder?

How Much Americans Actually Pay in Taxes

Second graph: share of pre-tax income vs. share of federal tax liabilities for five quintiles of income.

Note that this is for federal taxes, which are progressive; most state tax systems are actually regressive (because of high dependance on sales and property taxes) and so will even out this result a bit, making the share of liabilities a bit closer to the share of income. See this report for details.

So most people pay a little less a percent of their income than the average, while the very rich pay a little more. I don’t know whether you call that “fair”, but it certainly seems reasonable. At any rate, I don’t remember when market economics was ever concerned with “fairness”–what really matters is what tax rate is acceptable enough to rich people that they won’t leave the country and thereby reduce overall revenues.

Don C: Dang, Bubba got there first. I’d also recommend Perspectives on the Reagan Years by John Logan Palmer, The States, the People, and the Reagan Years by AFSCME (a good state-level account), etc. The fact that Reagan tried to cut social spending is not a wild conspiracy theory – he ran on a platform of cutting social spending, was ideologically opposed to social spending, and cut social spending.

Barbar: It’s a good question. I think that actually you can have both at the same time, i.e, that we needed debt to cover our consumption, and that we had capital that needed to be spent somewhere (indeed, if you take the position that the issue is an imbalance in wages/profits, the two are inherently linked). The question beccmes – what would have fueled economic growth if it hadn’t been massive debt? If incomes are stagnant and falling over time, you need something to make up the slack and cover the increasing inventories.

Don C: I am glad you asked “what should the average “minimum/living wage” be in America?) What is the percentage of a person’s income that you feel is “fair” to confiscate in the form of taxes?” These are important questions to answer, and progressives should enjoy answering them.

I would say that the minimum wage should be a living wage – i.e, enough to provide the basic necessities of life (housing, food, clothing, transit, etc..), the basics of economic security (health coverage, some form of retirement savings, child care, and so forth), and the potential for economic mobility (one’s own training/education, homeowning, enough savings to cushion against a sudden crisis, enough money for one’s children’s higher education). This site: http://www.livingwage.geog.psu.edu/ is quite helpful on this score. Take my state of California – it takes about $10.72/hr to make it as a single adult, and $24.62 to make it as a two adult/one kid household. I think that’s more than fair.

As for taxes, first of all, I don’t think of it as confiscation, but rather a purchasing of collective goods – roads, police, fire, schools, emergency health, and the like all cost money. I would say that it sort of depends on your income – being taxed at 50% of your income means destitution if you’re making $50,000 a year, but if you’re making $5,000,000 a year, I don’t see it as a burden because you’ve still got enough to live comfortably on. But thinking broadly, I would say that 0-5% is good for people earning within 150% of poverty, call it 10-15% for people within 250%, 15-25% for people within 350%, and upwards. Personally, I don’t have a problem with the old 90% top marginal bracket, because at that level of income, the cash is really only a way to keep score.

It’s certainly true that inequality has increased over the past 25 years, but the first graph does nothing to illustrate this fact: the scale means that even if incomes at the middle of the distribution had grown as fast as those at the top, the bottom line would still look flat compared to the top one.

You should either use a log scale, or let 1985 income = 100 for all groups, or plot growth rates of incomes rather than levels.

It’s always been my (casual, ignorant) observation that the underlying trend behind all the great social upheavals of history has been income inequality. Whenever there’s a great social injustice, the size of the reaction, from complete disinterest to violent revolution, I suspect tracks closely with income inequality. I know this is a terrible oversimplification, but I wouldn’t be surprised if social inequality tends to make societies less stable and more prone to revolt.

«I would say that it sort of depends on your income – being taxed at 50% of your income means destitution if you’re making $50,000 a year, but if you’re making $5,000,000 a year, I don’t see it as a burden because you’ve still got enough to live comfortably on.»

So you if someone stole $1,000 from someone making $1,000,000/y that would be less of a crime than if the $1,000 were stolen from someone making $10,000 a year? Aren’t citizen equal before the law?

And wouldn’t stealing $1,000 from someone highly productive who contributes $1,000,000 to the economy every year be more damaging than stealing the same sum from someone who hardly contributes to the economy?

People like DonC start from very different assumption than yourself; typically that high income is a reward for winning, low income is a punishment for losing, and that taxation is a way to rewards losers and punish winners, and to impoverish the nation, by stealing resources from those who contribute most to those who contribute least, and that since everybody deserve what they get, stealing from high income producers is as immoral or more immoral as stealing from low income parasites.

Indeed following that logic the moral solution would be to have taxes paid by lower income parasites to higher income producers, to ensure that resources be distributed in a progressive way, to those who create progress, not to those who hold it back.

There are very good objections to DonC’s assumptions: that paying USA taxes is entirely voluntary, and nobody is forced to give a cent to the USA government (I don’t for example), and that what he calls “welfare” is really an entirely voluntary insurance pool against misfortune, and that self-insurance is extremely inefficient even for the wealthy (think fat-tail distributions).

Matt/Barry: I’d side with Barry on this. Many political revolutions -the French and American Revolutions are the best case – don’t take place in either states or periods in which inequality is unusually high, or poverty unusually widespread. Rather, they tend to come right after a period of prosperity dramatically reverses, dashing expectations that had previously been quite high and raising fears of a sudden decline in station.

lemuel said: “You should either use a log scale, or let 1985 income = 100 for all groups, or plot growth rates of incomes rather than levels.”

It’s never been clear to me why people think comparing percentage increases in income is appropriate. If someone making $20,000/year gets a 1% raise, that gives them another $200/year, which is about 1 new television of improvement to their life. If someone making $20,000,000/year gets a 1% raise, that gives them another $200,000/year, which is much more than 1 new television of improvement to their life.

For general quality-of-life issues, comparing +$200 vs +$200,000 seems more meaningful than comparing +1% vs +1%.

It occurs to me that the word “inequality” doesn’t do justice here.

“Inequality” assumes some level of participation: we engage in some productive activity together, we divide the proceeds; if I get 80% and you 20%, then you probably have a reason to complain about inequality.

In this case, however, it looks like the bottom 80% do not participate in distribution of the proceeds at all, they are being paid a fixed amount, regardless of their productivity. This is not inequality, this is more like a form of slave labor.

Steven Attwell @ 21:

Your 2000-2007 statistics are throw-away, partisan fodder; this isn’t about Bush administration policies, it’s about broad trends over the last 40 years. I presume you don’t really need to be convinced, but anyone who does can easily find out that the real incomes of those who have gone to college have increased by about $10K/year over that period, while the incomes of those who haven’t have pretty much not changed at all, just as I asserted above. (The picture is even more stark when you look at total compensation, since it’s the college grads who get the good benefits packages.)

Your second paragraph simply illustrates precisely the problem I pointed out in the original post: it generates ever larger income numbers by looking at ever smaller slices of the population. To see the effect a change on broad-based measures of inequality like the Gini coefficient, you need to weight the income change by the numer of people it affects. That’s why a “mere” 20% real increase in the incomes of millions of college graduates can outweigh a 300% real increase in the incomes of CEOs. By choosing the ratio of the 99th percentile to 50th percentile as your metric, you have simply defined away the possibility that broader contributors to inequality can play a role.

Finally, I’m not sure what “redistribution” in the 1945-1973 period you refer to. Government spending as a fraction of GDP has been pretty constant for the entire post-war era, the wealthy contribute a larger fraction of government revenue after 1973 than before 1973, and government transfer payments represent a larger fraction of that spending after 1973 than before 1973. Perhaps you refer to top income brackets of 90% during that early period, but given what I just pointed out, it’s hard to argue that those higher brackets correlated with more redistribution.

I actually don’t think tax brackets had much to do with rising inequality, but someone who did could well argue that what happened was that the elimination of those high brackets incentitized high earners to work harder, creating more wealth for themselves but also more income for government to redistribute.

To put a point on it: the really important drivers of inequality are not the CEOs, but people like the readers and writers of crookedtimber.

Household incomes at the 80th, 90th, and 95th percentile have all increased relative to the median in the past few decades, with larger increases the higher up you go. For example, the ratio of the 80th percentile to the median was 1.7 in 1967 and 2.0 in 2003.

I think David Wright is correct to say that the 99th percentile is not the only number to track. Ideally there’d be some authoritative paper out there breaking down the causes of the inequality shift.

Market fundamentalists have to face the fact that their God has failed

As with all fundamentalists, their God Cannot Fail; It Can Only Be Failed.

David Wright
~”it’s about broad trends over the last 40 years. I presume you don’t really need to be convinced, but anyone who does can easily find out that the real incomes of those who have gone to college have increased by about $10K/year over that period,”

Not to put too fine a stroke to it, but could you possibly delve a bit deeper in the that pot of paint? Are you saying that broadly speaking that I, my wife and most of those we know [all with college educations] have gotten a raise of about 80k during the GW Bush tenure or a roughly 400k raise over the past 40 years [or a percent thereof]? While I do know some very wealthy folk, and have lived and been employed for those 40 years with degrees for the best part of them … I see neither as have happened, broadly speaking of course.

For informational purposes, in 2007, the average CEO compensation for America’s top 500 companies was $6.4 million; the top 4 clocked in between $100 and $200 million.

The increases in the top percentile are significant in overall terms.

The increases at the 80th percentile have also been significant (although less so) in overall terms.

MarkUp @ 27: Sorry, my attempt to be clear with units apparently backfired. If you earned $30K/year in 1975 and $40K/year in 2005, your income increased by $10K/year; the rate of increase was only $330/year/year.

Marc @ 25: I appreciate your diving in with real, quantitative analysis! There are a few factors that make a careful quantitative analysis difficult, for example: Our income data is on households, but our education data is on individuals. Income volatility is very high for the highest income groups, so the income of the 99th percentile ratio depends strongly on the period you normalize over. The fraction of the population that falls into the college-educated category has increased significantly over the time period under study, but the fraction of the population in the top 1% category remains… 1%. I’ll try to dig up articles which actually compute the Gini coefficient under different scenarios, but I think from these sorts of back-of-the-envelope calculations it’s safe and fair to say this: neither the top-1%-income-explosion nor the college-education-effect is negligible, and neither is overwhelmingly dominant as a cause of inequality.

As an aside, I don’t entirely understand your tax incentivization argument. If I am greedy and I get to choose how the rewards of our common enterprise will be divied up between us, won’t saying that the government will take 90% of what I get cause me to choose to direct more to myself, to make up for that loss? In any case, please note that I specifically wrote that I personally believe tax rates are a red herring. (Since income data is pre-tax, they are in any case second-order effects.) I was merely pointing out that Atwell’s tax ideas a double-edged sword.

Babar @ 32 makes an important point. If your concern is really about the kind of super-star compensation (celebreties, hedge fund managers) that garners press coverage, you’re really talking about the top-0.01%, not the top-1%. The top-1% includes a lot of “merely” highly educated, in-demand professionals (medical specialists, corporate tax lawyers). I think it will even include upper-middle-class coastal homeowners realizing their appreciation, because the homeowner capital gains exemption is “below the line” in U.S. tax law.

In the US, the debate about how much tax the rich pay has been hopelessly confused with the story of the Bush tax cuts. See for instance here. This old post is also useful.

So you if someone stole $1,000 from someone making $1,000,000/y that would be less of a crime than if the $1,000 were stolen from someone making $10,000 a year? Aren’t citizen equal before the law?

Taxing (or stealing, if you prefer) everyone by a fixed amount is what’s know as a poll tax. They’re very, very unpopular. The last person who tried to institute one was Margaret Thatcher; she lost power over the fiasco.

To continue your example, imagine if you taxed $1000 on someone who was making $100 a year. They’re now out of pocket for $900. That’s why poll taxes often lead to riots.

And wouldn’t stealing $1,000 from someone highly productive who contributes $1,000,000 to the economy every year be more damaging than stealing the same sum from someone who hardly contributes to the economy?

No, it isn’t more damaging. Let’s leave aside the naive assumption that productivity is proportional to one’s income; a cleaner is more productive than your average AIG executive. You are going to get less random property damage from taxpayers over a progressive or proportional tax than from a regressive tax. Losing $1000 from $1,000,000 leaves $999,000, and we could all do well with a spare $999,000. Losing $1000 from $10,000 leaves $9000 – making it harder to feed yourself, let alone your kids (and what about their uniforms and textbooks too?). That’s the sort of situation where bricks get thrown at tax offices.

Federal, state and local tax combined:

I’m not convinced we can or should do anything about income inequality, but we should make taxation a lot more progressive.

Income taxes are not the only taxes. If you combine ALL U.S. taxes — federal, state and local — together, then the percentage distribution of taxes to income is nearly FLAT, down to the top of the bottom quintile. Most people apparently don’t know this.

Total taxation should be increased progressively up to a top 50% tax rate. A nice, easy number that everyone can remember. This figure can be approached by increasing the top federal income tax rates, and closing the loopholes.

It won’t hurt economic growth, it never did — except in chalk on a schoolroom blackboard. Most members of the economics profession have been acceding for decades to the clownish psychology and propaganda of the thinktank op-edders claiming that taxation of the rich hurts innovation.

In reality, the reverse is more likely: the evidence shows that economic growth has not been harmed by government spending for 200 years (see Peter H. Lindert, Growing Public: Social Spending and Economic Growth Since the Eighteenth Century, vol. 1.)

RE: Bricks thrown thru windows@39: The last word on the art of taxation was offered by Cardinal Richelieu, who opined that the art of successful taxation was akin to the skill in the plucking of the Goose for it’s down feathers “with the minimal amount of hissing.”

I think the 90+% tax bracket/ incentive conundrum could be explained like this:

There is (and always will be) a small number of extremely avaricious (and unethical) individuals in any given population. These individuals analyze the social scene and find the vocations that would (under the circumstances) provide the best opportunity for them to achieve their objective. Currently, ‘business manager’ is certainly high on the list. These individuals are very difficult to compete against, and so people with genuine interest in business management have little chance to make it to the top.

Now, imagine that the society makes an effort to deincentivize these individuals. And not only from taking CEO jobs, but from polluting any legitimate profession: doctors, lawyers, stock traders, etc. This can be achieved by introducing something like a “maximum income”, or 100% tax bracket; and this would undoubtedly cause most of these asocial individuals (aka “top earners/producers”) to pursue illegitimate, criminal careers instead. Voila – suddenly everything is back to normal: managers manage, doctors heal, thieves steal. Everybody’s happy, case solved.

Looking at the Citizens for Tax Justice site referred to by the recent Economist’s View post I linked to above, I find a report called “Do the Rich Really Pay Over a Third of Their Income in Federal Income Taxes?” (.pdf). An extract (quoting a recent report by the IRS):

“The IRS data show that the federal income tax rates paid by the highest-income Americans have dropped substantially since 2000, largely due to cuts in the tax rates on capital gains and dividends pushed through by the Bush Administration. While income from work (salaries and wages) is subject to rates as high as 35 percent, income from investments (long-term capital gains and stock dividends) is taxed at only 15 percent.
The IRS report shows that in 2006 (the latest year for which data are available), the 400 richest income tax filers paid just 17.2 percent of their adjusted gross income (AGI) in federal income taxes. That is down from 22.3 percent in 2000, and is less than half of the top statutory income tax rate of 35 percent. Almost 65 percent of the income reported by those 400 taxpayers consisted of capital gains and dividends subject to the preferential rates”.

On the “Stay Tuned”-bit: if property increasingly is concentrated than the income of a happy few will always increase faster than is good for anybody, including themselves – high time that Soros/Buffett and the very rich liberals are barred from pontificating, & forced to loosen their more-leverage-than-the-US-President-grip on world economy.

As has been pointed out elsewhere, one simple explanation for at least part of the increasing household inequality over the last 40 years is that woman are working and now men and women both tend to seek mates who are doing roughly as well as they are. Suddenly household incomes of $100K vs $50K become $200K vs 100K, a 50K increase in household inequality without any one person earning more.

Does this ‘assortive mating’ become a matter for concern?

“Stealing” and “Fairness”.
If taxation is stealing, there should be no taxation. None. Everyone can “donate” or “contribute” what they think they need to to keep the roads, water, fire dept, police and military going. If you get in trouble with anything, good luck.
If we want “fairness”, then I want a shot at defining the rules. Breaking the rules is unfair. All men are created equal. You get zero dollars from any inheritance. We are all equal. That’s fair. You get zero advantage from the wealth of your family. We are all raised in a collective. That’s fair. We all get the exact same education, work the exact same hours and get the exact same pay. That’s equal, that’s fair.

Lane, it looks to like like, using the after tax after transfer numbers, median household income has gone from $35,435 to $46,699 over the period 1979-2003 (in 2003 dollars), an increase of $11,264, while median household income under the standard definition (which I think is what you are using) goes from $38,649 to $43,318 over the same period (also in 2003 dollars), for an increase of $4,669. In other words, using post-tax numbers suggests much stronger median household income growth over the relevant period. What am I missing?

My own person feelings – which aren’t very liberal at all – is that it is the latter case.

Before I comment any further SOV, I’d like to know at approximately what point you think non-economic factors become significant in salary determination: $50K, $100K, $250K, 1M+? (Personally, I think it’s probably true at the $1M+, but I’ve heard arguments that no man is really worth more than ~$60K a year).

Before I comment any further SOV, I’d like to know at approximately what point you think non-economic factors become significant in salary determination: $50K, $100K, $250K, 1M+? (Personally, I think it’s probably true at the $1M+, but I’ve heard arguments that no man is really worth more than ~$60K a year).

I don’t think there is a monetary set point; I’m merely pointing out that to the extent inequality is justified as a matter of cold economic value, few people have problems with it. It’s when value != compensation that a lot of people tend to get ticked off and that there is some sort of populist movement to redress the imbalance. And contrary to a certain brand of market fundamentalism, there are certain objective indicators that this is so. For example, if a person’s increase in productivity is 60% higher this year than last but their compensation remains flat, one could deduce that at some point this person is not being paid what they are worth.

For a specific job you mean? Could you be more specific? If someone produces 100 blorgs in year 0 and 160 blorgs in year 1, that would be a productivity increase of 160%. Not all jobs could be so bluntly assessed of course.

Not all jobs could be so bluntly assessed of course

Nicely understated. Matter of fact, I reckon that there is hardly any blorg in a modern economy whose marginal output can be ascribed to one person’s (or one factor’s) contribution. If we had a productivity-o-meter, this whole discussion would not be taking place.

You beg to differ, and yet it seems we agree after all.

My suggestion is that real-world production can be better modeled with a Leontief function, where factors are combined in fixed proportions and so the entire marginal product can be economically attributed to any of them. The division of output is then the result of a bargaining, or more broadly social, process.

I’ve never been quite comfortable with current executive pay levels, but I’ve also not been able to understand why owners and stock holders are willing to pay that much (and from what I have seen, it’s not just that the game is rigged by the boards, private owners seem to feel that their executives are worth similar amounts and that money is coming from their own pockets.)

I find it hard to believe that no company has succeeded in paying reasonable executive compensation and been rewarded for it without others following suit. Could the market have failed so thoroughly?

My own guess is that owners, stock holders, and those that companies must deal with to survive tend to use compensation as a metric for competence. Paying your executives what we would consider reasonable salaries signals to everyone that your executives are incompetent. (After all, if they were competent, they’d go elsewhere.) For businesses that depend on trading partners, that could be fatal.

One correlation that I have heard is that the compensation versus company size had remained roughly constant. It’s that companies are many times bigger than they used to be. Maybe you pay in relation to responsibility rather than to competence. Do supervisors of bigger school board earn more than for the same job in smaller school boards?

I don’t think it’s mysterious at all. For better or ill, the CEO can have a huge impact on a company’s profits.

I agree with this and the rest of your comment, but I think the past year has shown that CEO search committees could be doing a better job.

So: we can’t guarantee an efficient and rational market for CEOs, but we can guarantee an efficient use of their salaries if the government takes a large portion of it.

Lane, yep–somehow I must have written down the wrong number on definition 14 income, which led me astray. (In case anyone cares, $46,699 above should be $43,629.) Thanks.