Rich, Poor and Middle-Class

Mitt Romney when he said “I’m not concerned about the very poor…”

Today I want to react to my friend Karl Smith, to my cousin Mitt Romney, and to the part of the American electorate Mitt was trying to pander to when he said:

I’m not concerned about the very poor — we have a safety net there. If it needs repair, I’ll fix it. I’m not concerned about the very rich — they’re doing just fine. I’m concerned about the very heart of America, the 90-95 percent of Americans who right now are struggling.

I think Mitt has things exactly backward here. Fortunately, I have high hopes that he said it precisely because it isn’t true: that he is concerned about the very poor and the very rich, but has a lot of trouble connecting emotionally with the middle-class. So, since he might be our next president, oh may it be true that Mitt, in his heart of hearts, agrees with me when I say this:

I am deeply concerned about the poor, because they are truly suffering, even with what safety net exists. Helping them is one of our highest ethical obligations. I am deeply concerned about the honest rich–not so much for themselves, though their welfare counts too–but because they provide goods and services that make our lives better, because they provide jobs, because they help ensure that we can get good returns for our retirement saving, and because we already depend on them so much for tax revenue. But for the middle-class, who count heavily because they make up the bulk of our society, I have a stern message. We are paying too high a price when we tax the middle class in order to give benefits to the middle-class–and taxing the rich to give benefits to the middle-class would only make things worse. The primary job of the government in relation to the middle-class has to be to help them help themselves, through education, through loans, through libertarian paternalism, and by stopping the dishonest rich from preying on the middle-class through deceit and chicanery. 

With that statement in hand, it is easy to answer Karl Smith, whowrites this in response to my post “Why Taxes are Bad”:

If we agree that benefits reduce work effort, while taxes reduce utility then basic tax transfer mechanisms seem to offer us exactly the tradeoff we want.

Welfare benefits discourage the poor from working. However, the poor have a low marginal productivity of labor.  So, society loses little when poor people work less.

On the other hand rich people have a low marginal utility of income. So if we extract their utility by extracting their income society loses little.

However, when we redistribute money to the poor society gains a lot because the poor have a high marginal utility of income.

Isn’t this exactly the trade we are looking for?

On the part of the poor, my answer is “Of course we should help the poor more, and should not worry much about any resulting lost output by the poor.” However, we do need to be concerned that for many, a job is an important contributor to self-esteem, in a way that the poor do not always fully take into account when they let tax and benefit incentives influence whether or not they work.

As for the rich, as I write above, it is not so much their welfare I am concerned about, though that counts too, but getting incentives right to encourage them to work hard for the benefit of the rest of society. Among other things, the best of the rich provide a crucial decentralized guiding role for our economy.

But Karl’s point that benefits reduce work effort is a big problem when those benefits go to the middle class. The middle class make up the bulk of our society and produce the bulk of GDP. So if government benefits cause them to work less–for example, to retire earlier than they otherwise would–we all end up poorer.

Two last points: First, it will take me at least a hundred posts to develop, clarify and defend my views about taxation and related aspects of public policy. Second, many readers will be surprised that I think Romney might care about the poor (though not so much about the middle class) in his heart of hearts. This could be my familial bias talking, but I try to back up this view in my post “Will Mitt’s Mormonism Make Him a Supply-Side Liberal?”

Also inspired by a saying of Mitt’s:                  

“Corporations are People, My Friend”

Inspired by a saying of Barack’s:

“You Didn’t Build That: America Edition”

Dissertation Topic 3: Public Savings Systems that Lift the No-Margin-Buying Constraint

New York Stock Exchange

QUESTION: Miles, I am a 2nd year Phd student at the University of Athens,Greece and i would like to write a strong proposal to continue the Phd abroad (I don’t think you need details about why, just read the headlines!). I was reading your papers about precautionary savings and i was thinking to study the impact of social security on savings and investment behavior of the households. What do you think and where should I focus my attention? Thank you in advance!

I find this subject intriguing because I believe that the relation between healthcare systems, saving and macro performance is strong. The first two have been investigated extensively, but not for Europe despite the fact that the “experiment” is interesting due to the diversity between countries with respect to healthcare systems since data are relatively recent (thanks to S.H.A.R.E.). For the second pair I think that distorted saving affect the economy severely. Households savings are substituted by government and this could potentially affect the structure of the economy. For example some countries regulate the investing decisions of public institution and constrain to invest nationally and safely (usually national bonds!). Ideally i would like to focus in the most fruitful research topic and construct a theoretical model with heterogeneous agents and test it empirically.I understand that my questions are very broad and i need to narrow them.

ANSWER: In this area, I think there is a very interesting contrarian research project. Given even a modest positive equity premium (expected returns of stocks higher than expected returns of bonds), models of optimal life-cycle saving and portfolio choice suggest that–if only they could freely do margin-buying at the Treasury-bill rate and with no extra investment fees–young people should typically have what seem like very large investments in the stock market because the present value (total asset value) of their many years of future labor earnings is in the hundreds of thousands, or even millions of dollars, making substantial stock-holdings reasonable. Even at my age, if I knew how to do margin-buying conveniently at the Treasury-bill rate with no fees in order to effectively borrow to hold more stock than the total value of my retirement savings plan, I would do it, but I can’t. But government retirement savings programs could make this possible. Moreover, if the political flak could be survived, a public program is a way to overcome the fact that most people don’t understand the principle that they should integrate future labor earnings into their financial investment decisions.  (There is a lot of evidence to back up the fact that in the real world most people don’t understand this, even though in standard economic models, all of the “agents” serving are theoretical stand-ins for human beings understand this principle with perfection.)

As someone writing an economics dissertation, the political flak that such a program would entail is not your problem. Showing that it would substantially raise welfare if it ever happened would be very interesting. I think this is a research project that would intrigue people because it sounds so counter-intuitive, especially in the context of the current economic troubles, yet is so firmly founded in sound economic theory.

By the way, the idea that people ought to have large stock holdings even if the expected returns on stocks are only modestly above the expected returns for bonds is dramatically reinforced if one is willing to follow Raj Chetty in estimating risk aversion using labor supply behavior. I love this paper:

Raj Chetty, “A New Method of Estimating Risk Aversion”

American Economic Review, 96(5), 1821-1834, December

Using Raj’s numbers for risk aversion amounts to saying that only declining marginal utility is a sound reason for avoiding risk and that any aspect of apparent risk aversion that comes from departures from the standard expected utility theory does not deserve respect in thinking about the normative (prescriptive) question of how much risk people should take on in order to get higher expected returns.

I personally would find a job market paper that pursued this idea seriously and well quite exciting, and I don’t think I would be alone.  

This project would be computationally intensive. For style of paper, you might want to consider something like my paper “Portfolio Rebalancing in General Equilibrium” with Matthew Shapiro, Tyler Shumway and Jing Zhang or the partial equilibrium counterpart to that paper.

Is Taxing Capital OK?

The “Dark Satanic Mills” that Made England Rich

QUESTION

Hi Miles. As an English Literature Graduate with wider interests and intro Economics Classes, I wanted to ask you about something I read recently. Ostensibly, Joseph Stiglitz in his new book talked of a world where movement of labour was unimpeded, and cities and countries competed for the best workers; something that would be financed by taxing capital. This is deeply attractive to me, but I don’t know if it is possible, economically/demographically, even if political issues,permit. Thoughts? – workedspace

ANSWER

I haven’t read the Stiglitz book, so I will have to answer from general principles. Let me know if Stiglitz somehow has a way around the problems I will point out.

In simple economic models taxing capital has one of the biggest long-run negative effects on the economy of any tax. It looks OK in the short run, but with lower investment, the capital stock gradually declines. In this spirit you would be better off taxing land a la Henry George, since the amount of land won’t decline even if you tax it. But taxing the buildings on top of the land is like taxing any other kind of capital. (However, right now we tax houses very lightly compared to factories, so if it weren’t for the housing bubble’s aftermath, we would be better off taxing houses more and factories—which employ people—less.)

One way to tax capital some in a way that won’t hurt capital formation is to shift from labor taxation (such as Social Security taxes) to consumption taxation, since in the long run the shift to consumption taxation increases taxes on people who have the wealth to consume more than they earn. But in the shorter run, the shift to consumption taxation hits people who are temporarily having to consume more than they earn. Also, consumption taxation has a big life-cycle element. The biggest category of people who have the wealth to consume more than they earn is senior citizens. 

To return to the negative effects of capital taxation, the short-run temptation is much like the short-run temptation to have rent control. In the short run, the supply of apartments is inelastic, so rent control looks like a pure transfer from landlords to tenants, but in the long run rent control is a disaster because it makes it unattractive to build new apartments or even keep the old apartments in good repair. The stories of evil landlords not taking care of apartments that pop up under rent control are oh so predictable from economic theory.

At the beginning I said that in simple economic models, taxing capital has one of the most negative effects of any tax. That is true if the tax on capital is constant. If the government taxes capital now and promises never to tax it again, the story gets more interesting. In theory, forcing all companies to issue non-voting stock to the government worth 90% of a firm’s value would have no distorting effects, and so would be the perfect tax as long as people believed the government would never do it again. But if the government will do this once, what is to stop it from using the same logic to do it again? This is called the “dynamic inconsistency” problem.

Getting government institutions set up to block the recurrence of this ultimately self-contradictory logic behind taxing capital a lot now and promising never to do it again is actually one of the trickiest problems implicit in “Leveling Up: Making the Transition from Poor Country to Rich Country.” I think often here of the history of England. Confiscating accumulated wealth was always attractive to the king. Only the evolution of limitations on the King’s power to take accumulated wealth (the essence of the “one-time” capital tax) in the end allowed England to get the development of factories that helped it to become rich.

Note

This is the first published answer to a question using the “Ask me anything” button.

Update

There is a further Twitter discussion of this issue in this thread

. 

Health Economics

The caduceus, symbol of Hermes, Greek god of medicine

I am slow to post about health care because I don’t know the answers. But then I don’t think anyone knows the answers. There are many excellent ideas for trying to improve health care, but we just don’t know how different changes will work in practice at the level of entire health care systems.

Much of the political heat over health care reform has to do with the perception on both sides that the Affordable Care Act (Obamacare) is a move in the direction of redistribution. As a mode of redistribution it has many of the same issues as other modes of redistribution.  Redistribution is good, but when financed on a massive scale by the government, it can also be a budget buster. The extent to which this budget-busting aspect of a large amount of additional redistribution can be muted by extra efficiencies wrung out of the health care system is simply unknown.

An aspect of our public policy even before the affordable care act has been favoring health care expenditures relative to other forms of consumption.  In particular, people have long been able to pay for health insurance–but not most other forms of consumption–with pre-tax dollars. I think this can be justified by the fact that most of us are seriously bothered by thinking of others suffering without adequate medical care much more than we are bothered by thinking of others not being able to take family trips or having a small house or car. So it is worth something to us if others tilt their spending toward health care more than they would without any push toward health care spending in the tax system. As an example of a less subjective externality from health care, I think people’s psychological problems often cause them to act in ways harmful to their friends, extended families and coworkers, so I think it is appropriate for policy to tilt people’s spending toward any form of psychological care that can be shown to be effective at improving how people treat others around them along with whatever other effects it has. (Tilting should not be allowed to totally suppress price signals that indicate that some forms of psychological care require many more resources to provide than others.)

This principle of subsidizing what benefits others besides the one choosing how much health care to use is helpful in showing what forms of health care should not be favored. For example, plastic surgery for people who already look OK has at best mixed effects on how others feel. Am I misremembering the former Italian Prime Minister Silvio Berlusconi wanting to subsidize plastic surgery for the benefit of his own viewing pleasure? But those who are in social competition may feel worse off, and I think this externality is stronger than the Berlusconi externality. So, depending on the strength of different externalities, it may make sense for public policy to discourage plastic surgery for people who already look OK rather than encourage it. The ethical status of envy of others’ plastic-surgery enhanced looks–let alone the Berlusconi externality–is not an easy question, but at least one can say that the argument for using policy to tilt people towards spending on plastic surgery is muddy at best, and the default should be no tilt.

On wringing efficiencies out of the health care system so that we can hope to afford the large amount of additional redistribution in the Affordable Care Act, to me it seems crucial to have a great deal of experimentation rather than a one-size-fits-all approach. On the constitutional question of what the Federal Government can do and what States should be left to decide, Greg Mankiw refers to a previous Mankiw post saying that from an economic point of view taxes, subsidies and fines can all be equivalent. The 16th amendment to the constitution gives the Federal Government breathtaking power:

The Congress shall have power to lay and collect taxes on incomes, from whatever source derived, without apportionment among the several States, and without regard to any census or enumeration.

But what the Federal Government can do in relation to the States is not the same as what is should do.  Supreme Court Justice Louis Brandeis, in a dissenting opinion in 1932 said: 

It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.

This has come to be known as the “laboratories of democracy” principle, which I have always found very attractive. In the case of the Affordable Care Act, I believe that whether we are ultimately able to wring efficiencies out of the health care system depends on how much state-level experimentation is allowed. And that in turn is largely a matter of how the next President (whoever that turns out to be) interprets the Affordable Care Act. So even though I think it unlikely that the Affordable Care Act can be repealed, given the difficult designed in by our founders of getting any new legislation through Congress, it matters whether a President is elected who will give many waivers to states to try different experiments with health care. I hope that journalists–and others who get the chance to ask questions of the two major candidates–press them on this question of how freely they would give waivers for states to try various experiments if, as is most likely, the Affordable Care Act is not repealed.    

There is an obvious role for the economics profession in such state-level experimentation on how to deliver health-care. The government needs to ensure that there is adequate data collection in relation to these various experiments, and economists need to analyze that data. More generally, with health care spending at 17.4% of US GDP and rising, we need more economists working on health care issues than ever. In addition to current health economists redoubling their efforts, it is high time for economics departments around the country to give more prominence to health economics in graduate training than they have, so that there will be more health economists in the future. And I hope that where they reasonably can, empirical economists (and theorists) who do not now think of themselves as health economists tilt their research agendas toward figuring out health care. I stand by my statement that no one knows the answers for health care. But I hope someday that will no longer be true.  

Mark Thoma: Laughing at the Laffer Curve →

In my first post “What is a Supply-Side Liberal?” I wrote

I believe the harm to the productive performance of the economy caused by taxes and regulations is serious (though seldom serious enough that a reduction in taxes would raise revenue).

Mark Thoma’s post (title above is a link) shows that the weight of informed opinion is behind my parenthetical remark. I can verify that since I started econ grad school in 1982, having attended many economics seminars and having had many informal discussions with economists, I have never in person heard an academic economist argue that tax cuts raise revenue, with the possible exception of Larry Lindsey (Greg Mankiw’s and my boss when we were both section-leaders in Harvard’s Ec 10). Larry Lindsey argued that cuts in capital gains tax rates would cause investors to change the timing of capital gains realizations enough that cutting capital gains tax rates would raise revenue now–implicitly at the expense of revenue later, though he didn’t emphasize that.  

One reason that tax cuts don’t raise revenues is that the effect of taxes on GDP is itself complex, and can go either way. See my posts “Can Taxes Raise GDP?” and “Why Taxes are Bad.” If marginal tax rates can be cut both now and in the future it raises efficiency (a good thing) but it will typically make people feel richer as well, so that work hours won’t go up much, if at all (also a good thing).  

Update: Scott Sumner pointed out to me that the disagreement of economists with the statement

A cut in federal income tax rates in the US right now would raise taxable income enough so that the annual total tax revenue would be higher within five years than without the tax cut.

indicates that the overwhelming majority of economists don’t believe that tax cuts can raise revenue for Keynesian reasons either.  (And actually, disagreeing with this statement means not believing that the combination of supply-side and demand-side effects of tax cuts is enough to lead to an increase in tax revenue.) On the question of whether tax cuts can raise revenue for Keynesian reasons, see Valerie Ramey’s Powerpoint discussion of a recent paper by Brad Delong and Larry Summers.

Mark Thoma on Rainy Day Funds for States

Mark Thoma, Professor of Economics at the University of Oregon and Author ofthe Blog Economist’s View, which is the Center of the Economics Blogosphere

I have to apologize to many in the blogosphere for not having been a regular blog reader in the past and so not being aware of posts preceding mine that are clearly relevant to my posts and so deserve acknowledgement.  Unfortunately, writing this blog on top of all my other duties as an economics professor and in my private life also doesn’t leave me a lot of time for reading everything out there that I ideally should be reading.  So please do let me know of things out there that are highly relevant to my posts and I will make an effort to acknowledge them–assuming I agree that they are highly relevant.  This post is an acknowledgement of an article by Mark Thoma that is extremely relevant to one of my posts.  Mark let me know by tweeting about his prior article with @mileskimball included in the tweet. 

Back in October 26, 2010, Mark Thoma suggested in the Fiscal Times that the Federal Government help states out financially in return for states setting up rainy day funds.  (I have to apologize to many in the blogosphere for not having been a regular blog reader in the past and so not being aware of many things that people have said.)  This is very similar to my proposal in “Leading States in the Fiscal Two-Step."  So (assuming that, like Ben Bernanke, Mark thinks more fiscal stimulus is in order at this point) both Mark and I are calling for this kind of action.  One difference in our proposals is that I am suggesting that the Federal Government effectively require the states to repay the money given them now, and then go beyond that to accumulate positive balances in their rainy day fund.  Thus, my proposal will not add to the Federal Government’s debt in the end, while Mark’s will. 

Also, note that since my proposal works through changing the timing of Federal Medicaid Contributions, the Federal Government can easily do it unilaterally, and the effective rainy day fund requires no state legislative or executive action in order to come into existence. 

Avoiding Fiscal Armageddon

I believe the continued existence of our species is of great value.  Therefore it is worth paying a great price to reduce the chances of a nuclear Armageddon by even a little bit.  There is reason to believe that if Iran gets nuclear weapons, it will significantly raise the chances of further nuclear proliferation–which in turn will raise the chances of nuclear war further down the road.  Thus, to my mind, preventing Iran from getting nuclear weapons trumps any other foreign policy or economic goal right now.  Though we should fervently hope that through diplomacy the price is lower, we should be willing to pay whatever price is required, even if that price includes literal battle deaths as well as the figurative body blow of higher oil prices to an already ailing world economy. 

The surface of what was once called “our sister planet” Venus is not a pleasant place for human beings.  Scientists believe that Venus became so hot because of a runaway greenhouse effect.  We don´t know how likely it is that the Earth will go down this road, but the possibility is terrifying.  All of the other possible harms from increasing the atmospheric concentration of carbon dioxide–such as floods, hurricanes, famines, extinctions of many plant and animal species–pale in insignificance compared to even small increases in the chances of a runaway greenhouse effect that would truly transform Earth into Venus´s sister planet.  To my mind, the scientific and popular discussion about global warming should focus more on what we know about the possibility of a climate change Armageddon in the form of going down the Venus road and how we can know more, thanon the lesser horsemen of floods, hurricanes, famines, extinctions of plants and animals, and so on.

This graph of government spending as a percentage of GDP (from Jim Bianco) isn´t as scary as the other two pictures, but it only includes Federal spending, and absent decisive action, the worst is yet to come.  The reason the worst is yet to come is simple: the U.S. population is getting older as birth rates drop and the Baby Boomers (from an earlier era of high birth rates) reach retirement age.  Other than defending our country and acting as the world´s policeman, the big expenses of the Federal government are the programs intended to take care of older Americans: Social Security, Medicare, and the part of Medicaid that goes to pay for nursing-home costs of people who have used up their own resources.  Everything else the Federal government does pales in comparison to Defense and taking care of older Americans (although Obamacare could add a large enough government responsibility for the medical care of younger Americans to compete in this contest).  Thus, more older Americans means more expenses of the Federal Government.

As I have only begun to explain in my posts “What is a Supply-Side Liberal?”“Can Taxes Raise GDP?” and my current favorite post “Why Taxes are Bad,” the taxes needed to support high levels of government spending come at serious cost.  This is likely to continue to be true even if we try to be very clever (as we should try to be) at raising revenue with the least possible distortion.   (On one way to raise some revenue while actually reducing distortion, see “A Supply-Side Liberal Joins the Pigou Club” and “Henry George and the Carbon Tax.”)  So at the end of the day, after all our efforts to raise revenues in ways that are not too harmful to the many important objectives we put under the label “the economy,” there will be a limit to how large a fraction of GDP should be devoted to government spending.  

For the sake of argument, let´s suppose we decide that–except in temporary emergencies–that Federal, State and Local governments combined should spend no more than half of GDP.  How can we set up constitutional limits to achieve such a goal?  It won´t be easy.  As I wrote in “Leading States in the Fiscal Two-Step”:

… fiscal discipline is hard for governments—so hard that most governments are not capable of fiscal discipline without balanced budget requirements that are so inflexible they cannot easily handle economic emergencies.Given the unpredictability of real-world events, it is not possible to fully define what constitutes an economic emergency in advance as flexibly as would be desirable, but in the absence of hard and fast, but inflexible rules, governments have a temptation to chronically declare economic emergencies.

It´s hard, and maybe impossible.  But let´s try.  A balanced budget amendment won´t do it, since it would still be possible to tax and spend a large fraction of the (possibly smaller) GDP pie.   A limit on taxes won´t do it, since the Federal Government is good at deficit spending.   What we need is a rule that goes directly toward the goal: a direct limit on the fraction of GDP that is allowed to go to government spending. 

There are complications.  In particular, we need to (1) be able to get the rule passed, (2) make it hard to wiggle around the rule once it is in place, (3) be able to deal with emergencies and (4) avoid tempting the government to skimp on investing for the future.

(1) Getting the spending ceiling passed.  The Republicans have been successful at making a big deal of the periodic need to raise the ceiling on the national debt.  Suppose that the next time the debt ceiling needs to be raised, they said “We are willing to make increases in the debt ceiling automatic.  All we want in return is immediate legislation and passage by Congress of a matching constitutional amendment that limits government spending by all levels of government to less than half of GDP."  I think that the Democrats would lose politically by opposing this; if they opposed it, they would be admitting that in the future they planned to take more than half of GDP for government purposes.

(2) Making the spending ceiling stick.  My idea here is simple.  If government spending at all levels (Federal, State and Local) is more than 49% of GDP (according to a modified formula discussed below), the President of the United States has full power to cut Federal spending of any type in any way that he or she chooses down to the level that makes total government spending 49% of GDP.  If government spending at all levels is more than 50% of GDP, the President of the United States is required to cut Federal spending to bring total government spending down to less than 50% of GDP–and of course is allowed to go further down to 49% if he or she chooses. Since the members of Congress value their own power, they would be unlikely to actually pass a spending bill that went above the 49% level, since that gives away a lot of their power to the President.  So the likely outcome is that budget negotiations would proceed much as they do now, but Congress would never propose anything that pushed total government spending above 49% of GDP according to the formula.  The one exception is if they thought they didn´t have the self-discipline to do what needed to be done and wanted the President to take the political burden from them.  One reason to leave the President full flexibility on how to cut is that he or she would only be in that situation if Congress had given up on trying to control spending even though not trying to control spending themselves meant giving away a big chunk of their power to the President. 

3. Designing a formula that allows for emergencies.  Designing a rule that can deal with recessions, wars and other emergencies is complicated, but I think can be done.  The first step is to use five-year averages for a number of quantities that tend to go up and down with the business cycle.  I am going to give a list of things I would use five-year averages for.  These are going to be things that can move around even if Congress hasn´t done anything in terms of regular budget legislation.  The list: GDP, government spending on unemployment insurance and welfare, spending by state and local governments, interest payments on the debt (which are counted as spending), and military spending.  The five-year average for military spending would include only a cobbling-together of the most recent 60 months that were not during a period of declared war.  The five-year average for unemployment insurance and welfare and state and local spending would include only a cobbling-together of the most recent 60 months that were not during a period when the 3-month Treasury bill rate was below a .25% annual rate, which is meant to leave out periods of such great economic emergency that the Fed has already gone to its limit on reducing short-term government interest rates.   The five-year average for GDP would be the greater of (a) the average over the last five years and (b) a cobbling-together of the most recent 60 months when the 3-month T-bill rate was above .25%.   

There are several points to make about this formula.  In a recession (according to a standard story) having unemployment insurance payments and welfare payments go up somewhat helps to stabilize the economy.  These increases in spending are given some leeway by not being counted fully right away.  Also, I am hoping that previously putting in place a Federal Medicaid Contribution Prepayment and Escrowing program–as proposed in my post "Leading States in the Fiscal Two-Step”–would make state government spending go up in recessions.  This too would be given extra leeway by not being counted fully right away.  And unemployment insurance spending, welfare spending, and state and local spending during periods when the Fed had already brought interest rates down to near zero would be fully exempted from the calculation.  Similarly, a military spending spike in a given year would only be partially counted in that year, and not counted at all in the formula if during a period of declared war.  And the spending limit would be in relation to GDP as it was before a serious recession.  Finally, using the five-year average for interest payments on the Federal debt cushions the effect of any change in interest rates on how tight the spending ceiling is. 

Looking at the other side of the ledger–the possibility of having the purpose of the rule subverted–there are several points to make:

• It is not that easy to wiggle around the rule in the long run using the 5-year average since spending more in one year then tightens the effective limits in each of the following 4 years. 
• A period of economic emergency is effectively determined by the Fed (by having interest rates very low), not Congress or the President, so that does not present too much temptation.
• I think formal declarations of war are taken very seriously.  

There is one other feature that may not be obvious but is also important in providing discipline.  I am proposing to have the GDP numbers and the spending numbers be nominal amounts, which means that there is no inflation adjustment that Congress could tinker with.  Doing the five-year moving averages for all of GDP but for only some types of spending means that the spending ceiling would effectively be tighter if inflation is higher.  So this spending ceiling actually provides some discipline against Congress pushing the Fed toward too much inflation.  (I don´t think the Fed acting on its own would try to cause extra inflation just to tighten the spending ceiling.)   

One last issue with the formula is that one might worry that State and Local government spending will trend up so much that the Federal government can´t do its job.  The answer to that is simple: if that becomes a big problem, the Federal Government has the power under its authority to regulate interstate commerce to limit State and Local spending to some reasonable maximum. 

4. Capital Budgeting.  Having roads and bridges in key places is important for economic growth.  I don´t see a big problem with allowing the Federal Government to do the same kind of thing as local governments do with millages.  If a ten-year bond is issued to pay for a road that lasts longer than that, it is OK to count just the annual payments on that ten-year bond as government spending for the purposes of the spending ceiling.  The reason to go to the trouble to set up capital budgeting rules is to avoid giving Congress an incentive to neglect important investments.  But note that the spending ceiling still bites.  In each of ten years the money being paid for the road would mean the Federal Government is required to spend less on other things.  To avoid chicanery, I would make ten years the longest payment period allowed for capital budgeting, even for things that might last longer.  

Something that doesn´t require explicit capital budgeting, but is somewhat similar, is the way that Federal Lines of Credit (as proposed in my post “Getting the Biggest Bang for the Buck in Fiscal Policy”) would interact with the spending.  The key is to make clear in the rule that Federal loans (which Federal Lines of Credit provide to individuals in order to stimulate the economy) are only counted as spending when repayment of those loans is behind schedule.  Any loan that was being paid off on schedule would not be counted as government spending.  And a loan that fell behind in a previous year but got back on schedule this year would be netted out against other loans that fell behind.  

Though I know the word “bailout” has a bad odor these days, if bailing out banks ever again became necessary to stabilize the financial system, bail-out loans could be treated similarly: only counting loans in actual arrears on payments as having been government spending.  The risk that loans might not pay off in the future would not be counted as government spending. Even if the loans are ultimately going to go bad, the key here is to spread out when the government´s losses are counted as government spending, so that the government can do the bailouts if absolutely necessary.  There should be a big debate about bailouts, but I don´t want to tie the governments hands in a true emergency, so the spending ceiling I am proposing does not attempt to stop bailouts. 

Without decisive action, I fear that we will end up with government spending more than half of GDP.  I think it will be much easier to get a spending ceiling like this enacted while it still seems shocking to think of running more than half of our economy in one way or another through the government.  Avoiding fiscal Armageddon will not be easy in any case, but it will be easier if we set up good rules now to limit a possible torrent of spending in the future.

Leading States in the Fiscal Two-Step

[Note on the cartoon:  My technical and blog etiquette expert Diana Kimball tells me it is proper etiquette to acknowledge the origin of photos by linking to the source.  I won’t go back and do that on all of my earlier posts, but I wanted to give you the link to this one because I found it early on, and then was frustrated that I couldn’t find it when it came time to post “Leading States in the Fiscal Two-Step."  I love this cartoon.  It is much better than the cartoon I previously had of Uncle Sam dancing with Lady Liberty.  Here is the link.  From the site, it looks as if you can ask people online to draw pictures for you pro bono–something useful for every blogger to know.  Has anyone had any experience with this?]

In his article "The Federal Reserve Turns Left,” Bill Greider tells how the Fed is currently one of the institutions arguing most strongly that the Federal Government should spend more now.  Since the Fed has not departed from its long-term recommendation that the Federal Government spend less in the future, this amounts to recommending that the U.S. Federal Government execute what I will call a “fiscal two-step”: spending more now, while promising to spend less later.

I have serious misgivings about this recommendation.First and foremost, I think it is politically very hard for a sovereign nation to do this:temporarily higher spending is likely to get built into the baseline.Second, even if such a fiscal two-step as usually envisioned is executed as intended, it will lead to a path for the national debt that is considerably higher over a substantial length of time (with all the risks that entails) than an alternative policy I have put forward on this blog of Federal Lines of Credit (FLOC’s) now combined with National Rainy Day Accounts (NRDA’s) down the road.

The combination of FLOC’s and NRDA’s effectively encourages people to do the individual version of the fiscal two-step.By providing lines of credit, people are encouraged to spend now if spending now is attractive to them.Later as part of the FLOC program they are required to repay this debt, and later still, they are required to save a modest amount each month by payroll deduction (or for the self-employed, in their quarterly estimated taxes) to go into their National Rainy Day Accounts–money that they are only allowed touse in the wake of a recession or other declared economic emergency or in case of a bona fide, documentable personal financial emergency of high order.With the combination of FLOC’s now and NRDA’s later, the FLOC’s now would provide stimulus now; while later on, in recessions down the road, stimulus would be provided by releasing some of the accumulated funds in the National Rainy Day Accounts.(If this combination of policies is followed, and it is a long time before the next recession, FLOC’s might only be needed once, since releasing funds from National Rainy Day Accounts could provide the needed stimulus in the future.But an additional round of FLOC’s would always be available as a backup policy.)

To give an idea of the possible magnitudes in National Rainy Day Accounts, suppose each adult were required to set aside $20 per month in NRDA’s.Then if it was five years between the institution of NRDA’s and the next recession, compound interest would mean that each adult would then have a warchest of more than $1200 in their NRDA , minus the amount that went to take care ofdocumentable personal financial emergencies.That amount would be available for release to fight that future recession.

Why am I repeating all of this, when I said it already?I am recapping all of this as background for a corresponding policy proposal for the states, which has a non-obvious name: Federal Medicaid Contribution Prepayment and Escrowing. The idea is to have the Federal Governmentgive the states more money now, while in effect requiring them to repay it later, then to build up a balance that would legally belong to the states, but remain in escrow until a recession, when some or all of the funds would be released.(Also, in situations where under current policies, a state might be bailed out–or otherwise rescued–with Federal money, the money in a state’s Medicaid Escrow Account could be released, so that at least the state was being bailed out or rescued with its own money.)

Interest would be assessed on the prepaid amount and earned by a state on the escrowed amount at the 3-month Treasury bill rate.Up to some target maximum for the Medicaid Escrow Accounts (specified in the original legislation), the interest would be paid as an addition to escrowed funds, but if a state’s account reached its target maximum, the interest would be paid to that state as unencumbered funds that the state could spend at will.The reason for insisting that the Federal Government pay the states interest in an unencumbered way once some predefined target maximumis reached is to reaffirm the states’ property rights in their Medicaid Escrow Accounts.Without ongoing reaffirmation of the states’ property rights in these Medicaid Escrow Accounts, the Federal Government would be very strongly tempted to confiscate the money at some future time in order to deal with its own fiscal problems.The Federal Government could always unilaterally cut its transfers to the states, but it is important that it be discouraged from penalizing a state for being especially thrifty with its Medicaid Escrow Account or for being slow to beg for a bailout or rescue.

Let me say with the utmost clarity that there is no real connection between this program and the purposes of Medicaid itself.I am imagining the rules of the Medicaid program itself to be totally unchanged.The only thing that would change is the timing of when the Federal Government pays its share, as defined under current law.

The main reason I am envisioning this program happening as a modification to the existing flows of Medicaid money from the Federal government to the state governments is simply for the sake of political plausibility.I think it is much more likely, politically, that the Federal Government would change the timing of its Medicaid contributions than that the Federal Government would explicitly set up the equivalent of FLOC’s and NRDA’s for the states.

The second reason for working through a modification of the existing system of Federal Medicaid Contributions is that this approach makes it easier to work around the balanced budget requirements in state constitutions.I do believe that macroeconomic stabilization is an appropriate use of the Federal Government’s power to regulate interstate commerce, and the Supremacy clause in the Constitution of the United States should allow the Federal government to override balanced budget requirements in state constitutions in any case.But with a policy of Federal Medicaid Contribution Prepayment and Escrowing, this can be handled simply by saying in the legislation

“This change in the timing of Federal contributions for Medicaid shall not be construed as borrowing or as saving when any State or Federal court interprets balanced budget requirements that states impose upon themselves.”

Thus, the policy would actually have the effect of tightening existing state balanced budget requirements in non-recessionary times, while loosening those requirements during recessions.

Along these lines, let me emphasize one more time the importance of the Federal Medicaid Contribution Escrowing down the line, as well as the Federal Medicaid Contribution Prepayment now.It is important for states to have the funds in those Medicaid Escrow Accounts to fight future recessions.And it is just as important to keep the states from overspending in good years as it is to encourage them to spend in bad years.

Why do I think the fiscal two-step will work for states, when I don’t think it will work for the Federal Government? Because fiscal discipline is hard for governments—so hard that most governments are not capable of fiscal discipline without balanced budget requirements that are so inflexible they cannot easily handle economic emergencies.Given the unpredictability of real-world events, it is not possible to fully define what constitutes an economic emergency in advance as flexibly as would be desirable, but in the absence of hard and fast, but inflexible rules, governments have a temptation to chronically declare economic emergencies.

The solution I am proposing here is to let a higher government(facing different political incentives) enforce the fiscal discipline in a way flexible enough to deal with recessions or other genuine economic emergencies. I think a government can lead in a fiscal two-step, and I think a government can follow in a fiscal two-step.  (Though they may do so clumsily, stepping on each other’s toes.)  But it takes two to tango, and two to dance the fiscal two-step.Almost no government can lead itself in such an intricate dance.

Henry George and the Carbon Tax: A Quick Response to Noah Smith →

The title is a link to Noah Smith’s post “Carbon Taxes Won’t Work.  Here’s What Will.”  

I have to be quick because I have proofs to read for my new AER paper with Dan Benjamin, Ori Heffetz and Alex Rees-Jones.  One downside of blogging is that your coauthors know what you are doing instead of working on your joint paper with them.  

Forgive me, Noah, if I am missing something in only skimming your post, but I wanted to say that a carbon tax or a gasoline tax isn’t just about global warming.  I like the national security argument for lowering the price of oil that, say, Iran gets, as well as the global warming argument.  I second you in advocating scientific research as the most effective way to address global warming.  But the revenue from the carbon tax can help pay for this research and other important government functions.    

To the extent that oil or coal in the ground is in inelastic supply, increasing taxes on oil and coal in many countries around the world at the same time is somewhat like Henry George’s land tax that came up in the conversation when Clive Crook told Michael Kinsley about this blog, as reported in Clive Crook’s post “Supply-Side Liberals.”  Note that if oil or coal is in inelastic supply like this, the before-tax price of oil and coal might go down so much that the after-tax price of oil and coal doesn’t change that much.  If so, the quantity used might not change that much either.*  So the direct benefit on the global warming front of a carbon tax might be small.  But the benefit in raising revenue to pay for research to solve the global warming problem could be huge.  

Overall, relative ineffectiveness at slowing global warming would lead to a smaller optimal level of a carbon tax, but the revenue side alone is pretty favorable, as long as the government doesn’t give away all of the potential revenue by using “Cap and Trade.”

*Note:  I have to give credit where it is due for this idea that worldwide carbon taxes will push down the price of oil or coal enough that there might not be much effect on use.  When I was a visitor at the Center for Economic Studies in Munich last Summer, I learned this at lunch from one of the other visitors there, who was unsuccessfully (I think) trying to persuade one of the graduate students in Munich to work on it.  I hope that someone has worked on this idea or plans to work on it.

A Supply-Side Liberal Joins the Pigou Club

I try not to oversimplify in my posts, but I confess to oversimplifying on occasion in the titles of my posts.  The number of words it is reasonable to put into a title–and a title’s function of intriguing people enough that they might actually read a post–make me willing to make that compromise.  

In the title to my last post “Why Taxes are Bad,” I am referring to income, earnings and consumption taxes.  There is a big exception to the rule that most taxes have bad side-effects: Pigou taxes.  Pigou taxes are taxes that discourage people from doing too much of things that have bad consequences for the world that don’t fall entirely on the person making the decision.  Pigou taxes are one of the closest things there is to a free lunch in economic policy: a way to finance the many important things the government does–including enabling the government to defend our country and help the poor, build roads and bridges and foster scientific research–while discouraging things that should be discouraged (at least discouraged a little bit).   

The most important Pigou tax on the table (at least among economists) is a carbon tax.  Because one of the big benefits of Pigou taxes is to finance important government functions without causing harmful distortions, I am unwilling to consider “Cap and Trade” as an equivalent to a Pigou tax on carbon emissions.   In theory, if the carbon emissions rights in “Cap and Trade” were auctioned off, it would be equivalent to a carbon tax at some level (though I think less likely to be at the right level than if the carbon tax rate is set directly), but in practice, most of the potential government revenue would be given away to carbon-emitting companies.  That would mean either that important government functions would suffer, or that we would have to use distortionary taxes–with all the attendant costs detailed in "Why Taxes are Bad“–in order to finance those government functions.  The best can be the enemy of the good, but for now I am holding out for an actual carbon tax.  (If a comprehensive carbon tax is not politically possible, an increase in the gasoline tax would be a reasonable start, but I am not at all convinced that an increase in the gasoline tax is any easier politically than a comprehensive carbon tax.)   

In any case, I am hereby officially joining the Pigou Club.  For more about the Pigou Club and the logic behind Pigou taxes, see Greg Mankiw’s post ”The Pigou Club Manifesto.“    

Why Taxes are Bad

In “Miles Kimball, the Supply-Side Liberal,” Noah Smith summarizes my post “Can Taxes Raise GDP?” with this passage:

Many people are familiar with the fact that people work about the same  amount whether taxes are low (as in the 2000s) or high (as in the 1960s). Miles agrees with this, but points out that high taxes hurt people in a different way, by making them feel so poor that they have to work more, and thus depriving them of leisure.

This makes it clear to me that I didn’t get some of my point across in “Can Taxes Raise GDP?” since what Noah is saying only applies to the case when the government is wasting the money.  What if the government isn't wasting the money?  Let’s take the case of the government redistributing money from the rich to the poor (which is a big element of what is going on in Medicaid and Medicare, for example).  The first thing that happens, as Karl Smith emphasizes in “Welcome Miles Kimball: Now Set the Record Straight on Taxes” is that the poor receiving the money work less:

benefits lead people to work less, a point I have long tried to make. 

This is a big effect.  And if it means that fathers and mothers don’t have to work second jobs to make ends meet and can spend more time with their children, this is a good thing, even though it makes GDP go down.  (One of the many problems with using GDP as if it were the measure of how well a society is doing is that child-rearing by parents doesn’t get counted in GDP.)

So far so good, but what about the side effects of taking the money from the rich to give to the poor?  The “Occupy Wall Street” movement and its spinoffs paint a picture of rich people who got rich by some kind of chicanery or evil–a picture embedded in the negative connotation now carried by the phrase “the 1%.”  But the typical case is someone who got rich–or whose parents got rich–by providing a useful service.  And I am not just talking about the wonderful services provided by someone like Steve Jobs or Oprah Winfrey. For vividness I need to use famous people as examples, but before their current fame, in the relative anonymity that most rich people live in, Duchess Kate’s parents rose to become part of the “the 1%” in England by providing people with party supplies and decorations.   There were many parties that were better because Carole and Michael Middleton provided these supplies and decorations.  Aside from the rare cases (think Bernie Madoff) in which someone gets rich by pretending to provide a service they are not really providing, or by means of some other evil–we all benefit from having the rich work hard to provide us with those services. Also, many of the rich provide jobs to employees.  

Now let’s look at the bad side-effects of tax distortions. To do this, we need to compare distortionary taxes (such as income taxes, labor earnings taxes or consumption taxes) to taxes on the rich that do not depend on the level of their income, earnings or consumption.  If we could tax the rich in a way that did not depend on their income, earnings or consumption, they would fulfill their obligation to contribute toward helping the poor in part by working harder to make more money to pay the taxes.  This would benefit all of the people who receive the services they provide.   And they might hire additional employees to work with them in the extra time they are devoting to work.  

By contrast, distortionary taxes such as income, earnings or consumption taxes push the rich toward getting the money to pay their taxes out of their existing income rather than in part by raising their before-tax income to soften some of the hit to their after-tax income.  We are all then worse off as a result: the rich because they would rather work harder than take the tax hit entirely as a reduction in after-tax income, and everyone else who would have benefitted as a customer or employee from the extra services and jobs they would have provided.  

So why don’t we use non-distortionary taxes that don’t depend on income, earnings or consumption to raise the taxes to help the poor?  It is because it is not easy to tell the rich and the poor apart without looking at income, earnings or consumption.  

In principle, if we could look at genes, say, to figure out who is inherently blessed with the talents to be rich, we could avoid distortions.  This approach raises a host of troubling philosophical issues.  But it is taken seriously in the academic literature.  See for example my University of Michigan colleague Joel Slemrod’s paper with Kyle Logue: Genes as Tags: The Tax Implications of Widely Available Genetic Information.  At this point, I am not willing to recommend such a gene-based approach.  But without some new approach, we are stuck with identifying who is rich by looking at income, earnings or consumption.  So we use taxes that depend on income, earnings or consumption, with all their attendant distortions. 

My overall message in this post is given well by Noah’s summary of my first post “What is a Supply-Side Liberal”:

Taxes bad, redistribution good.  

Some messages are worth repeating.

National Rainy Day Accounts →

To back up the idea that there may be even better policies than FLOC’s, here is another proposal from my paper “Getting the Biggest Bang for the Buck in Fiscal Policy” in the  section: “3. Household Finance Considerations.”  

… in principle, national lines of credit in times of low demand could be superseded in the long run (at least in part) by a modest level of forced saving in times of high demand, with the funds from these “national rainy day accounts” released to households in time of recession (and also perhaps in the case of one of a well-defined list of documentable personal financial emergencies).  

In the long run, I like National Rainy Day Accounts much better than FLOCs.  But National Rainy Day Accounts seem politically harder to me than FLOCs, because National Rainy Day Accounts require action in advance of a crisis instead of after the crisis hits.  

The title of this post is a link to the paper.  

Can Taxes Raise GDP?

I appreciate Scott Sumner’s post, “A Kindred Spirit,” and Karl Smith’s post “Welcome Miles Kimball, Now Let’s Set the Record Straight on Taxes.”   Karl Smith raises the issue of whether taxes can raise GDP and rightly points out that a substantial income effect means that taxes can raise GDP.  Indeed, it is a standard result in Real Business Cycle Theory that once prices adjust, extra lump-sum taxes to finance, say, an expensive war, will raise GDP by making everyone feel poorer so that they work harder.   A simple lump-sum tax is a tax that falls equally on everyone in dollar terms:  say every adult needs to pay $5000 per year.

Continuing to think about very simple models and government spending on something like an expensive war, what about a consumption tax that is a certain percentage of everyone’s consumption?  On the one hand this makes people feel poorer so they want to work more, but on the other hand, what someone can buy with an extra hour of work is less, so they want to work less.  The standard view is that these two effects will roughly cancel each other out.  So the amount people want to work–and thus GDP in this simple model–will stay about the same.  Matthew Shapiro and I take exactly this standard view in our paper “Labor Supply: Are the Income and Substitution Effects Both Large or Both Small?”   The basic argument for the standard view is that to households, a consumption tax looks a lot like a wage cut.  And we have a lot of information about what higher or lower wages do to desired work hours.  Among people who have to live on their own wages, there is surprisingly little difference in how many hours people want to work based on whether they have high wages overall or low wages overall over the course of their lives.  

Scott Sumner replies to Karl Smith, saying among other things:

It’s true that measured GDP might not fall if the money is wasted (as poorer people tend to work harder.)  But that’s not much of an argument for big government.

Scott Sumner is absolutely right that any unnecessary government expenditure is to be lamented.  Robbing people of their leisure time as well as of the wherewithal to spend on household wants and needs for no good purpose is a crime.  

But suppose the war really is necessary–or that the President has decided on the war and that as his economic advisor the only thing you get to decide is how to pay for the war.  What would be the best way to pay for the war if everyone in the economy had the same income level so that poor vs. rich weren’t an issue?  

With everyone at the same income level, a lump-sum tax is fair.  It raises GDP.  The lump-sum tax leads everyone to take some of the hit from having to finance the war in lower consumption and some of the hit from having to finance the war in less leisure time.  Less time away from work leads to higher GDP as more hours at work make it possible to produce more.  Guns are counted in GDP as well as butter, and gun production goes up more than production of consumption goods such as butter goes down.  

By contrast, the consumption tax with its tax distortion pushes everyone to take all of the hit from having to finance the war in lower consumption, and none of the hit in less leisure time.  Production of consumption goods such as butter goes down so much that GDP stays the same even though production of guns goes up.  This is inefficient.  The production of guns is the same in both cases.  But it is less painful to cut back some on leisure and some on consumption (what happens with the lump-sum tax) than to take the whole hit in consumption (what happens with the consumption tax).  

So the bottom line is that, other than a concern with redistribution, a distortionary consumption tax to finance a war that leaves GDP the same is less efficient than a non-distortionary lump-sum tax to finance a war that has GDP go up to help make the war effort possible without sacrificing as much consumption.  

The broader message is that taxes can be bad even if they don’t lower GDP.

What is a Supply-Side Liberal?

As an Economics professor, thinking about public policy is a big part of my job, both in teaching and research.  The work of the ivory tower has given me some distance from the rough-and-tumble of daily political debate, but has called on me both to face the enduring dilemmas of public policy and to identify areas where technical solutions are available, but not generally understood.  

As for areas where technical solutions are available but not generally understood, one of the most important is in stabilization policy.  It does not seem to be generally understood that there is no shortage of powerful tools to revive both the U.S. economy and the world economy.  This is true despite (A) short-term interest rates already being close to zero in the U.S. and many other countries and (B) most countries not being able to afford to add much to their national debt.  I will post on this point soon.   

Among the enduring dilemmas of economic policy the most important is the conflict between efficiency and equity.  In calling myself a supply-sider I am saying that I believe the harm to the productive performance of the economy caused by taxes and regulations is serious (though seldom serious enough that a reduction in taxes would raise revenue).  In calling myself a liberal, I am saying that in addition to an attachment to the liberty, limited government, constitutionalism, and rule of law emphasized by Classical Liberalism,  I hold to a view based on both classic Utilitarianism and contested elements of modern economic theory that, generally speaking, a dollar is much more valuable to a poor person than to a rich person, and that therefore, there is a serious benefit to redistribution that must be weighed against the serious distortions caused by the usual methods of redistribution.     

Economists have identified two numbers that are central in governing the size of distortions caused by taxes and the benefits of redistribution: 

1. Tax distortions are governed in important measure by the the consumption-constant elasticity of labor supply.  The consumption-constant elasticity of labor supply measures how much less workers want to work when what they earn is taxed, but the tax revenue is recycled back to them in one form or another of government benefit they can get regardless of how little they work.  Matthew Shapiro and I argue in our paper “Labor Supply: Are the Income and Substitution Effects Both Large or Both Small?" that the consumption-constant elasticity of labor supply is large.  Even leaving aside the decision of whether to work or not and just focusing on how many hours to work, we found a consumption-constant elasticity of labor supply equal to one and a half.  
2. The benefits from redistribution are governed by what I will call the degree of inequality aversion.  In a research project that began in 2005 but is still ongoing, Fumio Ohtake, Yoshiro Tsutsui and I put some extra questions on the University of Michigan Survey of Consumers (the survey behind the Reuters/University of Michigan Consumer Sentiment Index).  We asked the people answering the survey first "It is often said that one thousand dollars is worth more to a poor family than to a rich family.  Do you agree?” Over 90% of everyone agreed.  Then we went on to ask them questions such as this:  "Think of two families like yours, one with half the income of your family and one with the same income as your family.  Which would make a bigger difference, one thousand dollars to the family with half your family’s income or four thousand dollars to the family with an income like yours?“  More than half of everyone answering the survey said that $1000 would make a bigger difference for the poorer family than $4000 for a family at their own income level.  As analysis ably assisted by Daniel Reck and Fudong Zhang confirms, this implies a degree of inequality aversion above two.  An inequality aversion of two would mean that if you double someone’s income the value of an extra dollar will drop to a quarter of what it was.  So $4000 to the family with twice the income looks like $1000 did to the family with the lower income.  The short summary is that most people believe that dollars are worth a lot more to the poor than to the rich when they are asked in a context not immediately connected to public policy.  But the public policy implications of this belief are dramatic when coupled with a view that making a net positive difference in people’s lives overall (added up across people) is a legitimate goal of public policy.    

Perhaps because of cognitive dissonance, it is common for people to either believe (a) that tax distortions are serious and redistribution is of questionable value OR (b) redistribution is valuable and the distortions induced by taxes are small. My belief is that (c) tax distortions are serious AND redistribution is valuable.  That makes me a supply-side liberal.