Prof. Bryan Caplan

Econ 321


Weeks 8-9: Taxation and Redistribution

I.             Taxes and Redistribution: The Basic Facts

A.           There are widespread misconceptions about the numbers on taxation and spending.  Let's start with some basic facts for 2007.  (

B.           For the federal budget in 2015, expenditures are comprised of roughly:



Social Security




Domestic Discretionary




Net Interest


Income Security




Other Retirement/Disability




Offsetting receipts


C.           Main facts to note: payment for the old add up to 41% of the budget, over twice spending on defense.  Payments for the poor come out to something like 18%.

D.           For the federal budget in 2015, revenues are comprised of roughly:



Individual Income Taxes


Payroll Taxes


Corporate Income Taxes


Excise Taxes/Customs




E.           Main facts to note: most taxes come from the items you see listed on your paycheck - income taxes, social security taxes, and Medicare-type taxes.

II.            The Leaky Bucket: The Deadweight Costs of Taxes and Redistribution

A.           Taxes and redistribution take wealth from some people and give it to other people.  That's pretty obvious, and there's no need to study economics to appreciate it.

B.           What's not obvious: The deadweight costs of taxation and redistribution.  In addition to transferring wealth, they also destroy some wealth in the process.

C.           The leaky bucket: in the process of transferring wealth, some "slips out," benefiting no one.  (Ice cream in the desert analogy makes the same point).

D.           Landsburg on "Why Taxes Are Bad"

E.           How can wealth simply be destroyed?  Many ways. 

1.            The effort of preparing tax forms, along with accountants, tax lawyers, etc

2.            Production foregone because of taxes

3.            Production foregone because of redistribution

4.            Diversion of effort into less productive - but less taxed - lines of work

5.            Producing things people value less (like medicine) instead of things they value more (like vacations).

F.            Basic idea: A tax that can't be avoided ("lump-sum taxes" or "head taxes") merely transfers income.  A tax that can be avoided will have deadweight costs because people change behavior to do so.

III.          Labor Taxation and Marginal Tax Rates

A.           Taxation of labor income is a basic part of the U.S. tax code.  As income rises, your assessed tax liability rises too.

B.           Key question: When you earn $1 more, how much more in tax do you pay?  If the answer is $1, you have a 100% marginal tax rate; if the answer is $.25, you have a 25% marginal tax rate.

C.           What are marginal federal taxes, and what are the cut-points?  Here they are for 2016 for single filing status:

Min $

Max $

Marginal Rate






















D.           Of course, you pay more than just the federal income tax.  You also pay SS tax, state income tax, etc.  Adding up all of them (and appropriately adjusting for deductibility!) tells you the critical question: If you work one more hour, what do you earn after taxes?

IV.          Leisure Subsidies and Marginal Benefit Reductions

A.           The government also subsidizes leisure by paying people who have little or no income.  Standard forms are welfare, unemployment insurance, and SS.

B.           Analytically, welfare-type programs are surprisingly similar to income taxes.  Two aspects:

1.            Give people, say, $500/month if they have $0 income.

2.            REDUCE their welfare payment 1:1 if they earn anything greater than $0.

C.           The initial payment makes it feasible to live without working.  The greater its size, the fewer people work.

D.           The 1:1 reduction feature leaves no incentive to work more than zero.  So if you go on welfare, you don't work at all.

E.           Bottom line: standard welfare programs first increase people's wealth, then raise their marginal tax rates to 100%.  Both discourage work.

V.           Policy and Labor Supply: Income and Substitution Effects

A.           So how do government tax and redistribution programs affect the quantity of labor supplied?

B.           Since tax laws apply throughout the economy, not merely isolated sectors, we need to think in terms of Aggregate Labor markets.

C.           From the point of view of workers, proportional labor income taxation (a "flat tax") is equivalent to a decline in Aggregate Labor Demand.  They get paid proportionately less for each hour of work.

D.           Does this necessarily reduce hours worked?  Surprisingly, no. 

E.           In Aggregate Labor markets, you have to think about both the income and the substitution effects.  Higher taxes reduce the return to work; but they also make people poorer, discouraging the consumption of everything - including leisure.

F.            Assume - as before - that income and substitution effects balance out, so Aggregate Labor Supply is vertical.  Then proportional labor income taxation has NO effect on total hours worked! 

1.            Absurd?  What would you do if the tax rate were 95%?

G.           Still, on reflection, the assumption of perfectly vertical labor supply may be too strong.  This may be sensible for prime-age males, but it overlooks some less obvious channels, such as:

1.            Female labor supply.  Married women in particular pay a lot of attention to their after-tax earnings when they decide whether to stay in or re-enter the labor force.

2.            Retirement age.  People nearing retirement age may be more likely to stop working as tax burdens rise.

3.            Others?

H.           Progressive tax systems - where the marginal tax rate increases - are much more likely to reduce hours worked.  Even with roughly equal income and substitution effects, they can reduce hours worked. 

I.             Hard to graph, but intuitively simple: Progressive rates let people earn enough to be comfortable, but then tax them at ever higher rates on their last hour of work.

1.            If female labor supply and retirement age is sensitive to proportional taxation, then they will be even more sensitive to progressive taxation.

J.            What about redistribution?  Recall that this raises recipients' income AND (progressively) raises their marginal tax rate.  This can be decomposed into two effects:

1.            Higher tax amounts to a reduction in ALD.

2.            Money not to work reduces ALS.

K.           Some have argued for simply abolishing welfare due to these effects. 

L.            A more moderate proposal has been the "negative income tax."  The essential idea is to reduce the marginal tax rate on welfare recipients below 100% to leave them with an incentive to work. 

VI.          Policy, Compensating Differentials, and Human Capital Acquisition

A.           While labor taxation probably doesn't have a large effect on the quantity of hours worked, it probably has big effects on the occupations people enter.

B.           Key feature of tax codes: You pay tax on income, but not "fun."  Thus, the higher taxes get, the more people will choose jobs for their "fun," rather than their usefulness to others.

C.           More generally, you generally do not pay tax on "non-cash income" such as free meals, coffee, etc.  (Though there are some legal limits - on parking to take one example).

D.           Suppose everyone received equal pay so long as they worked.  Everyone would then do what they loved, regardless of whether anyone else liked it.  There would be millions of actors, athletes, professors, etc., but few that any wanted to watch.

1.            Employers in this example would try to attract more productive workers with enormous non-cash benefits. 

E.           In my view, the shift into fun and non-cash income is the biggest real-world effect of income taxation.  It is particularly harmful that the most talented people face the highest marginal tax rates, and thus the weakest incentive to apply their abilities in a socially useful way.  

F.            If foregone time is the only cost of human capital acquisition, then proportional taxes don't affect it.  Why?  You get less, but also lose less.

G.           But human capital acquisition does fall if:

1.            Taxes are progressive

2.            Schooling is costly or unpleasant

H.           This effect may take time to reveal itself for life-cycle reasons.

VII.        Rationales for Redistribution

A.           Rationale #1: Redistribution as a return on investment.  For the largest program, SS, people supposedly get money because they previously contributed to the program.  They are just being paid a "return on their investment."

B.           Problems:

1.            If people really want to invest, they can do it on their own.

2.            Actual returns don't match contributions very well.  The first recipients of SS got a windfall; present recipients get a below-market return.

C.           Rationale #2: Redistribution as insurance.  Another story is that these are "insurance" programs.  People may not actually benefit from them, but they are assured that if they get sick, lose their job, etc., they will be cared for.

D.           Problems:

1.            If people really want insurance, they can buy it on their own.

2.            Premiums and benefits rarely adjust for risk like a real insurance policy.  The rich, for example, are extremely unlikely to go on welfare, but pay more to support these programs than the poor.

E.           Rationale #3: Egalitarian redistribution.  A third account is that redistribution deliberately aims to make poor people better off by making rich people share with them.

F.            Problems:

1.            Programs that benefit the elderly actually don't do this.  Why?  Because the rich live longer than the poor on average, so they wind up collecting more money from SS and Medicare.

2.            More importantly, if this were the real reason for redistribution, none of it would be spent on the relatively poor people in the U.S.  It would go to absolutely poor people in other countries.

G.           Rationale #4: Externalities.  Redistribution reduces crime, begging, and so on.

H.           Problems:

1.            Are the elasticities even close to high enough to make this a good idea?

2.            Will the elderly turn to crime?

VIII.       Programs Big and Small: The Old Versus the Poor

A.           Most redistribution focuses on the elderly: SS and Medicare amount to 35% of the budget.  The American poor get about 13% of the budget.

B.           Egalitarian arguments cut against old-age programs for demographic reasons: the wealthy on average out-live the poor by over a decade.

C.           Moreover, if people wanted to make investments or buy insurance, they could do so on their own.

D.           The real argument for old-age programs is mostly paternalism: "People aren't rational enough to save for their retirement, so we must force them for their own good."  But:

1.            Why force foresighted people who are planning for their future to participate? 

2.            Isn't lack of foresight in large part a product of paternalism itself?  Spencer quote.

E.           Egalitarian arguments also cut against real-world poverty programs, since they help relatively poor Americans, not absolutely poor foreigners.

F.            Both kinds of programs have important incentive effects. 

1.            Old-age programs distort retirement decisions. 

2.            Poverty programs affect not only work incentives, but are also probably the key to high teen pregnancy.

G.           Much of the money spent on the old and poor is for health care, which probably does little to benefit them considering the cost. 

H.           This is particularly clear for the old: Health care for the elderly is very expensive, but at best slightly lengthens what are probably the worst years of your life. 

I.             The same basic argument works for the poor.  They value health care less than the rich because they have more pressing priorities.  Imagine: If you were earning $10,000/year, how much would you want to spend on health care?

IX.          Redistribution in Reverse: Immigration Restrictions

A.           Actual redistribution looks more like "tribalism": it's not about helping the poor, but "taking care of your own" even if it means harming foreigners.

B.           Probably the best example: many favor immigration restrictions because people are "coming here to collect welfare."

1.            A simple compromise would be to give immigrants "second-class citizen" status: eligible to work but not collect welfare.

C.           Some frankly complain that immigration should be stopped because it hurts wages for low-skilled Americans.

D.           Either way, the idea is to help relatively poor Americans at the expense of absolutely poor foreigners.

X.           Why the Standard View of the Welfare State Is Wrong

A.           The "standard view" of the welfare state: there is a trade-off between compassion and efficiency.  The most compassionate policies would fully take care of the poor, but these would have severe efficiency costs.  Real-world policies try to strike a reasonable balance.  Life was terrible back in the 19th century before the welfare state existed; only "mean," and "uncaring" people could prefer it to what we have now.

B.           This is wrong on several levels.

C.           First, most of the welfare state is about helping the old, not the poor.

D.           Second, the help for the poor goes to relatively poor Americans who are already quite fortunate by global standards.

E.           Third, the goal of "helping the (American) poor" is probably the main justification for immigration restrictions that greatly harm poor foreigners.

F.            In the 19th century, people had to fend for themselves, but anyone was free to move to the U.S. and try their luck.  Policy was far more "compassionate" then than it is now, all things considered.