Bryan Caplan  

Conversational Evidence

More Classic Jeffrey Friedman... Conversational Evidence on Tax...
I've long argued that economists should take more stock in introspection.  Now Karl Smith argues that economists should take more stock in conversation:

[F]or example if you asked someone - why didn't you move to San Francisco or conversely, why did you move to Raleigh. Plenty of people will say, the cost of living. Indeed, it's a meme. Everyone knows that its cheaper to live in the South and this is a reason you might consider living there when you other wise wouldn't.

This is a strong clue that prices might be influencing folks location choices. And I mean clue. Its not definitive but it should make your ears perk up and say, "hmm is what is going on here what everyone thinks is going on."


There are a ton of phrases and terms for not working because of the welfare state:  "Put in my time"; "Hanging in there 'til 63"; "Mooching off the system"; "Welfare queens"; "Mailbox money" etc.

This is a thing that people talk about. That is a clue that it probably exists or at least that they mean to be talking about something. What that is, we would want to check and be sure.

Compelling.  But I think Karl is too hasty to use conversation to question the impact of taxation on behavior:

Causal empiricism just didn't support the notion that taxes are a big deal.

The reason is that people aren't talking about them in a way that they would talk about something that was a big deal. Oh, they complain sure but that's not what I mean...


When you ask why didn't you take that job at the bank do people say, well you know the tax bracket thing?

There is not even an easy way to describe not working because the after tax income made it not worth it. If there are no terms for it, then that's a clue people aren't talking about it, which in turn means that they aren't likely doing it.

Now, maybe we could have missed something and people are talking about it but in some sort of subtle way or in a way where they are confusing causes. Sure, I think those things are real possibilities. However, then someone should be able to find those phrases or terms. To my knowledge no one has.

Two points:

1. I often hear people say that doing some more X to make extra money "isn't worth it, especially considering Y."  Common Y's include transportation costs, childcare costs, and yes, taxes.

2. Proclaiming your responsiveness to taxation tends to come across as arrogant: "Oooh, I'm doing so well that I don't even bother.  Especially in my tax bracket."  People muffle their conversation accordingly - especially when they might be talking to people who don't face comparable tax incentives.

Either way, though, I applaud Karl's methodological open-mindedness.  You'd be surprised how much you can learn about economics just by listening to people.

Comments and Sharing

COMMENTS (40 to date)
Bill Woolsey writes:

I have always considered this a useful test of Ricardian equivalence.

How often do you hear about people talking about a change in their estate planning because of the current budget deficit. Man, after that Republican fillibuster, I went out and bought a CD. The grandkids will need some money to pay the taxes.

Eric writes:

I think introspection is at least a decent guide to what the introspecting individual might do, but the problem I have is that the introspecting individual is often not the marginal individual that determines economic events. The rise in unwed motherhood since the war on poverty took effect was a surprise to many people since they introspected and thought that an additional small paycheck wouldn't affect their choice to marry or have children. Unfortunately, they weren't the marginal individual. And as unwed motherhood became less stigmatized, the location of the marginal individual moved closer and closer to the average individual.

I've seen the example (on this blog, I think) of someone introspecting to figure out the effect of interest rates on demand for cash and saying, in effect, that a small change in rates wouldn't affect how much money they carry in their wallet. I strongly suspect, though, that people with money in their wallets are not the marginal decision makers on cash holding when interest rates change.

One final example: It surprised some people that no-fault insurance caused more accidents. Introspection says that I already have enough incentives not to crash my car, why would this change my behavior? Ultimately, though, I'm not the marginal individual. The marginal individual was the 21 year old reckless male who moves from just barely choosing to stay off the roads to just barely choosing to drive.

Bottom line: Incentives usually produce the theoretical consequences, often in surprising ways, even if introspection suggests they should not.

aretae writes:

It's a sizeable part of any discussion of homeschooling.

Roughly...homeschooling involves one parent not working (much). This would normally be a massive cost to homeschooling families without smaller children. Once you include taxes (the big thing) and other considerations (all, individually smaller), the cost differential vanishes. High marginal taxes are a major reason why homeschooling occurs. At least, it has been covered a lot in the last 20 year I've been talking about it.

Mark writes:

But why should I listen to people if they're out of my bubble? :>

Michael Hamilton writes:

I think I've heard other people discuss not working because of marginal tax rates:

"My father is taking all of us to Thailand for a month."


Eric writes:

I almost forgot. I think the reason you rarely hear conversations on marginal tax rates affecting behavior is because most people (the 20th-95th income bracket) are not the marginal people. These are people who work salaried jobs for 40ish hours a week and have very little day-to-day control over hours worked. They can't, for instance, choose to work 32 hours for 20% less pay or 50 hours for 25% more for the most part. Therefore, their behavior is to maximize their negotiated salary regardless of marginal rates. And their marginal rates are reasonable (sorry hard core libertarians :-). They aren't the marginal people.

There are three types of marginal people (I think). Low pay hourly workers, whose marginal rates approach 100%. They talk (I think) all the time about how an extra 50 cents an hour might cost them their section 8 housing allowance.

A second type is high-paid workers with consulting income. These people can make quick decisions about whether to work an additional 100 hours (spread out over a few months) in exchange for additional pay and face marginal rates where it matters. I recently joined this category and I guarantee that I decide whether or not to take a job based on my after-tax income. I do have a recurring (marginal) job that I take about 60% of the time. I'm outside the Social Security cutoff, and if I were made to pay an additional 12.4% self-employment tax, my acceptance of this job would probably drop to 0.

The third type is small business owners, who both make decisions regarding their costs and benefits, as well as whether to be a small business owner at all. I have less contact with this group, but I'd be surprised to discover that they didn't both talk about tax effects and act on them.

Kevin writes:

I hear people talk about decreasing their production due to taxes all the time. It usually comes in the form of some kind of semi-retirement where they work less or work a lower-stress job for less money. The people who talk this way can all afford to actually retire if it doesn't work out, and they do come across as arrogant as you say. I suspect they don't talk openly about this with strangers.

Eric writes:


I agree that it seems crass to talk about it openly. I felt weird typing the section:

"I do have a recurring (marginal) job that I take about 60% of the time. I'm outside the Social Security cutoff, and if I were made to pay an additional 12.4% self-employment tax, my acceptance of this job would probably drop to 0."

on a public forum and didn't use my full name. But it's still true. Greg Mankiw got all kinds of ridicule for admitting that he would change his behavior in response to taxes.

Alex J. writes:

Taxes were and are a major part of my wife's decision to not work very much until our child is old enough for kindergarten. We would have to pay a nanny and her taxes from our after tax income. My income puts us into a high enough bracket that we wouldn't be making much money off of her working, especially once you factor in the loss of flexibility from having both of us work.

Rick Hull writes:

There is a threshold to whether a particular topic merits conversation. Thus, it is clear that topics which meet this threshold are "important" to humans. But there are plenty of "important" topics which do not meet this threshold. I submit that the conversational test does not tell us much about such topics.

i.e. conversation implies important; no conversation does *not* imply not important

Telnar writes:

One bias in conversational evidence is that it tends to focus on things which are easier to change. A runner who would go faster in lighter gravity isn't likely to discuss that (ignoring special cases like travelling to an elevated stadium to break a record), while a runner who would go faster if he could find better sneakers has an incentive to discuss that problem since there's a good chance someone else has solved it.

Similarly, a worker whose hours are limited by lack of a good place to have dinner nearby has more reason to bring that up than a worker whose hours are equally limited by marginal tax rates (completely aside from the incentives from wanting to avoid bragging about one's financial situation).

Yancey Ward writes:

Smith's part on taxation is just utterly confused. Taxes are an embedded cost in everything, and income taxes are an embedded cost of every job decision. Also, he is completely ignoring the marginal decision maker in the one case while listening to them in the other (the re-locator, for example). Seriously, he has never heard in conversation someone complaining about marginal tax rates? Talk about being selectively deaf.

TomB writes:

In the DC area, many people list "lower taxes" as one reason they move out of DC and into Virginia. I would suspect people make similar statements in other cities that span jurisdictions. I also have heard of high-income earners moving to states like Florida to avoid state income taxes altogether.

I hear people talk about waiters preferring to get tips in cash, as opposed to being left on a credit card, because then they don't have to report the tips as income. Tax avoidance also is part of the reason why some restaurants are cash only.

You also hear people talk about how IRAs, 401(k)s and other tax-advantaged retirement savings accounts are good investments precisely because it lets them avoid taxes.

And you hear about people getting married before the the end of the year because then they can file jointly and wind up in a lower tax bracket.

So you do hear people talking about higher taxes mattering.

Seth writes:

I agree that I do hear people say, "it's not worth it." That's exactly what I thought when I was reading Smith's excerpt.

But, I also notice people don't often talk explicitly about opportunity costs, but they frequently act as if they exist.

Sara writes:

I'm a stay-at-home mom with three small kids. My husband's income is quite good, but not amazing (especially since we live in an expensive state and have three children.) My MTR is about 45%. Between that and day care I would have to make about $70,000 a year before I would take home $1. Not worth it. So I'm home with the kids. BTW, I've had this conversation with many many SAHMs around here and almost all of them are home because of this. Now it turns out I like being home and probably wouldn't go back anyway now, but that's not true of many people I've talked to. Ironically, with the rise of assortative mating, its often the highly educated, accomplished women who marry men with high enough incomes to put them in this predicament, so we're probably selecting for the drop out of some of our most productive workers. Maybe Karl Smith should talk to more moms.

Robbie writes:

There is actually a simple phrase that people use to describe tax influencing how they arrange their affairs: "for tax reasons".

I rent out the apartment I own to some tenants, and I lease a different place to live in, due to the strange combination of Australia's income and capital gains tax laws. When people ask me why, I say, "for tax reasons", and it is perfectly well understood and no more questions are asked.

I don't go into the details (and no one expects me to) because of the usual social norm that financial affairs (like how much you earn) are private.

Most people I speak to on the topic are very aware of the sharply lower marginal income tax rates in Hong Kong and Singapore compared with Australia, and they figure prominently in our conversations about the desirability of working there. "Income tax is 15% in Hong Kong!" is a perfectly normal, common and acceptable thing to say.

liberty writes:

"For tax reasons" is a huge one -- especially re: shifting money overseas, moving state-to-state, working less, moving money from cap gains to dividends earning accounts, changing business type, etc.

As for marginal rates in basic income brackets, there was SO MUCH talk of it at the lower end that eventually the government had to take notice and switch to the EITC which makes marginal rates negative, to make up for the high marginal rates of in-kind benefits (housing, food stamps, Medicaid, etc).

Some include:

"I'd be worse off if I worked, since I'd lose all my bennies."

"If I reported my earnings, I'd be screwed."

etc. .. Who cares how they say it? they all said it.

liberty writes:

And btw this is common discussion in the UK as well. Everyone knows that most of the people on the dole could get a job, but a job would have to pay fairly well to be worth losing the dole check and any other tied bennies (like housing benefit) plus the hard work, etc. This is common knowledge. "Not worth it" is the primary phrase. "It's not worth it to work in any job paying less than 20k/yr." This is all about marginal tax rates - whether negative tax is removed or positive tax applied. Everyone is aware of this and it is commonly discussed in the media.

On the upper-income scale, it's commonly discussed that people move their money around "for tax reasons" or "to avoid the higher tax rate" or they simply "don't bother" when the rate is high. This is commonly discussed by left and right on the radio and television - that people "change their behavior" when the tax rate is too high. As in the US the highest marginal rate was once in the 90 percent range, but recently even the 50% tax bracket is being discussed as not worth keeping because it brings in no more revenue than the 40% one, considering all the behavioral changes.

It's absurd to think these things aren't being discussed. Somebody is simply not listening.

Philip L writes:
The reason is that people aren't talking about them in a way that they would talk about something that was a big deal.

But since the return to individuals for participating in the political process is very low, why would they talk about it?

It's not like voting is rational to change it -- there are too many other people voting to make an individual count.

Wouldn't this incentivize people to discuss things that are more apt to their control?

cmprostreet writes:

I thought this sort of thing was why the phrase "Take-home pay" is usually used. It seems odd that in a search to find a phrase or euphemism to discuss marginal tax rates, this example didn't come up.

charles austin writes:

Isn't this just a variant of Pauline Kael's observation that she doesn't know how Nixon won because no one she knew voted for him?

Jeez, just how insular can you get?

[broken url removed--Econlib Ed.]

BCT writes:

I'm in the same position as Sara - well-educated stay at home mom etc in an affluent town full of the same. All these incredible women pour their smarts and energy into the schools and the community. Whenever politicians go on about 'the gap' between high achieving public school districts and low achieving (mostly poor, urban districts), realize among all the other factors, there is an army of talented SAHMs driving and supporting those high achieving schools.

bjhamill writes:

Ditto to Sara. I am a SAHM of 4 kids 8 and under. And I too had to make a career decision based at least in part on taxes. After childcare and taxes, I took home about 17 cents on the dollar. Literally. My husband's an economist and he does math for a living. B/c we're "rich" ( i.e. can afford something more than a crack house in the suburbs of DC) we don't get any of the child tax credits or child care expense deductions. Basically, I was working to pay uncle sam and my nanny. And my base was close to $100K. How is that not enough to make it worth it? It's unfortunate (bizarre? backwards? stupid?)that the current tax code discourages highly educated professional women from working when they have children.

JohnR writes:

It's not just the amount of tax, it's the bookkeeping and reporting requirements as well. I am an author and I will be giving a talk to about 100 people in the neighboring state of New Jersey next month. They asked me to bring some of my books to sell, since many would be interested. I checked with the NJ tax people about how to pay the sales tax for this one-time event. They said I had to register with the state and set up quarterly sales tax reporting. Since it was only for one night, I could dissolve my registration afterwards with another form once the taxes had been paid. And here's the best part: the registration costs $125! Even if evey person there buys a book, it just isn't worth it. So I lose money, the state loses revenue, and the people lose the convenience of buying on the spot. It's a small matter in the great scheme of things, but shows how taxes are a drag on even minor economic activities.

Chester White writes:

I may not talk about this stuff, but I have been thinking about it for 30 years.

I am a stay-at-home dad. My wife works, makes $160K. If I were to get a job, we'd have to get someone to watch the kid. I'd have to get another car. I'd have to pay more for gas and insurance. I'd have to buy a bunch of clothes and get them cleaned. We'd eat out more because we'd be too tired to fix food sometimes.

We'd be "rich," according to our d!psh!t president, and we'd be getting up to the highest brackets. I'd pay 15.2% or whatever it is in FICA/Medicare. State taxes. On and on.

I'd be in the 65+% effective bracket (minimum) all for the "joy" of killing myself 50-60 hours a week. The higher I go in income, the higher the wife's marginal bracket goes as well.

And we'd be pilloried anyway by every pundit and Demo politician and dumb@ss 99%er for not "paying our fair share." We'd soon find ourselves paying 75% for God's sake.

No thanks. I went Galt years ago.

Juliemarg writes:

Karl must never talk to high paid commission salespeople. All the top sales reps at AT&T and other places I've worked knew exactly how taxes would effect them.

If you are an employee whose income has wild swings from one pay period to the next, you are hugely impacted by tax brackets. A check for $10,000 is taxed as if you are paid $260,000 per year (26 pay periods x 10,000) even if that big check is an anomaly. People would roll their sales in to future pay periods to reduce the tax bite.

I'd argue that it was better to max income each pay period because you wouldn't be tempted to slack off in the next period because of banked earnings. I argued that in the end, you'd have made more money at year end my way, but I never persuaded anyone.

Argentum writes:

My wife and I find ourselves in the exact same situation as many here have already described. I make a decent salary with consulting and rental income on the side to boot. My wife is better educated than I am and could likely get a better paying salaried position, but as others have said, by the time you add up child care, another car, gas, insurance, etc... plus the time away from the family, what's the point?

I see no point in us busting our collective @$$es for what in the end is an additional $10K-$20K on top of our current income.

Screw it. I'm with Chester and I too went Galt some time ago.

Stephen writes:

I propose the following experiment: Have Congress pass and the President sign a law that ends as of January 1, 2013, all Roth, 401k, 403b, 457b and any other plan for tax-advantaged contributions to market accounts; then, record the effect on monthly flows into companies like, say, Fidelity. You might even pick up a conversation or two in the meantime.

Milo writes:

About 30 years ago while doing my preliminary chat with my accountant in March, I saw that the "bloodsuckers" were taking more than half of what I took home. NY State Income taxes, NYC income taxes, NJ income taxes, FICA, unemployment, a part of my group health insurance, etc. I also had to count in commutation, lunches and sometimes dinners. So I quit! With my new found freedom I found I could deliver coffees and lunches for a local diner, (using their car), for tips and a free lunch. Sort of my own way to go "John Galt" and do my part of starving the beast. Honestly, I miss doing what I did as I really liked it and was BTA at it. WTH. It beats being a slave.
Oh, and regarding Estate Planning: If the money runs out before I do, I can vote democrat so as to get some of other people's money.

h writes:

Eric's points above are insightful. As an example, I've noticed that my personal demand for milk seems to be perfectly inelastic: my wife tells me how much we need to buy each week, and I buy that amount regardless of the price. But in the aggregate, it is indisputable that milk demand is downward sloping: higher price means lower quantity purchased.

Also, Chester White (above) and Sara (above) make an important point. There is a big range of income at which net wages are negative or near zero. So it's not just a question of whether net income increases with a spouse entering the labor force, but whether the increase in net income per hour worked for the entering spouse is large enough to compensate for the loss in value of home production plus the value of leisure (per hour worked).

The combination of taxes and loss of income-contingent-government-benefits is an important factor. Even if you don't see it in your own behavior, it is important in the aggregate.

Todd writes:

[Comment removed pending confirmation of email address and for crude language. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Economic research, based neither on conversation nor introspection but, rather, on data before and after the 1986 tax law, suggests that Sara is correct. The biggest effect of marginal tax rates is on women married to high-income men. There may be more recent research, but I wrote about this research in 2000:

jamesbbkk writes:

There is a perverse cultural result of progressive taxation where people that complain about high taxes are treated as unsympathetic whiners because they supposedly have so much. By design, one might suppose. I, for one, deeply resent being burdened greater than others. I think it is immoral and creates all kinds of other perversions. Poor performers should be poor in wealth; it is merely natural.

Taxation of worldwide income and asset seizures of those who try to leave the onerous tax system of the US are also immoral and unethical policies which make us vassals of the state simply due to luck (or not) of location of birth but which apparently have popular support in the press and populace (tells us a bit about both groups I think).

There is indeed a term for responding to progressive taxation and related wealth confiscation by reducing productivity and income. It is: "going Galt." I also once knew a tax preparer who was told on a number of occasions by working customers that they did not take better jobs offered to them due to losses of tax credits. There may not be jargon for this, but such choices are made.

Dave writes:

Haven't seen anyone post how marginal tax rates influence U.S. military member residency choices. Or in other words wonder why a disproportionate number of military claim residency in Texas and Florida?

When I came on active duty I met with the finance officer as part of in-processing and he asked me what my state of residence was for tax purposes. I must of hesitated because he then pulled a binder off a shelf and said it might help. It had the tax rates of all the states. He told me I could choose Virgina, my home of record, Utah since I spent the past year there in school, or Ohio, the location of my new assignment. Ohio had the lowest rate then, so I became an Ohio resident. When I later was transferred to Texas, with no income tax, I became a Texas resident and remained one the rest of my career.

One might also check to see if marginal tax rates influence where military tend to retire. While I haven't studied that issue, I would be surprised if it is not a factor.

PJ writes:

At the other end, I retired early at 55 and worked at a p/t job for low wages but high satisfaction. Loved my clients (students). When I at 61 got a new mean boss, that satisfaction index hit rock bottom and I went to SS to get an estimate. I would make more at 62 than working, so I retired at 62!

I would do private work just akin to my previous job, but now Paypal must report more to the IRS, which means I get stuck with paperwork galore and the self-employment tax, etc. So far, I have decided not to do it. I do an occasional temp job for fun and spending money.

JayMan writes:

@Sara and the other SAHMs here:

If what you describe is true, and that taxation allows for otherwise potentially high-earning (i.e., high-IQ) women to stay home and raise children, then that is a good thing. Tell me, if you were working instead of being SAHMs, would you have as many children? Sara, as you mentioned, women in this type of situation are high-IQ/high work-ethic themselves thanks to assortative mating. It would seem then that, paradoxically, high taxation serves the eugenic purpose of encouraging fertility among high IQ individuals. This is actually an argument for the tax system as it is.

JoeFromSidney writes:

I'd like to say a word in favor of well-educated stay at home mothers. Yes, they would be productive in the economy, but is that the most important use for their talents and abilities? Why turn your children, your most precious investment in the future, over to someone of less talent and ability (otherwise they wouldn't be working as a nanny) while entering the labor market and paying taxes at your husband's marginal tax rate?

I claim that no matter how much money an educated woman can make in the labor market, she's doing much more long-run good raising her children to be productive citizens. We should encourage more such behavior.

Floccina writes:

Wive's of high earners often will say that will not work because it is not worth it. Men less often.

Floccina writes:

And BTW those wives do work, just not in the taxed economy.

m00tpoint writes:

Add us to the list of SAHM's primarily due to tax rates. My wife likes working. But when her job went away, we sat down and did the math. Her $50K job was netting us about $1.50/hr after taxes, transportation, the extra meals out, after school care for our 10 year old, etc. And the more progressive the tax rates get, the worse this will become.

Eat your hearts out, liberal feminists ... if you like women in the workplace, you might want to consider pushing for a flat tax.

Comments for this entry have been closed
Return to top