The rich, and the upwardly mobile

Matt Yglesias recently linked to a study by Cristobal Young and Charles Varner, which showed New Jersey’s “millionaire tax” (actually a half of millionaire tax) had little effect on migration.  Here’s the technique they used:

To further probe the impact of the new tax bracket, we develop a difference-in-difference estimator. For this, we exploit the fact that there is a plausible control group of high-income earners not subject to the new tax “” those earning $200,000 to $500,000 per year (Saez, Slemrod and Giertz, 2009). For this group, the new millionaire tax has no effect, and thus should have no impact on their migration patterns. In short, we use households in the 95th to 99th percentiles of the income distribution as a control group for households above the 99th percentile.

I don’t know how much effect the tax had, but I can’t have much confidence in the results of a paper that relies on assumption that 30-something doctors, lawyers, businessman, and entrepreneurs making $400,000/year would be indifferent to a tax increase on people making $500,000.  How many people do you know in that income bracket who don’t aspire to eventually earn even higher incomes?  Indeed, I’d expect the effect to be stronger for those making below $500,000, as they’d often be younger and not yet tied to a particular area (via kids in school, spouse with good job, etc.)

In my view taxes tend to have weak short run effects (as people are often settled in a particular area and don’t want to move), and stronger long run effects, as businesses deciding where to locate might choose areas with better fiscal regimes for their employees.  But many other factors also matter.

Places without many amenities cannot get away with high income taxes.  That’s one reason why lots more affluent people move to Texas than neighboring states with income taxes.  But state income taxes tend to be fairly low, and of course those lacking income taxes may have higher taxes in other areas.

In addition, some states have “captive” residents due to perceived amenities, such as California’s climate and NYC’s cultural attractions.  They can impose a 10% top rate and still hold on to many wealthy residents.  Indeed I plan to move to California in a few years, despite the fact that I much prefer Texas’s fiscal regime, and even more so its land use policies.

So the variations in state income taxes probably don’t have a major impact on population flows, but they may have a bigger impact than these studies suggest, especially in the long run.  My hunch is that states know this, and know that they can’t get aways with income taxes that are far outside the norm for other states with similar levels of amenities.  Indeed if tax regimes really didn’t matter, it would be pretty difficult to explain why New Hampshire’s population has grown much faster than any of the other New England states (even given the fact that it is slightly closer to Boston than the other New England states.)



17 Responses to “The rich, and the upwardly mobile”

  1. Gravatar of Jon Jon
    14. February 2011 at 08:54

    1) People need to work in NYC. That’s where the jobs are, especially the jobs that pay $500K. NYC has its own income tax, and NY state itself is the second highest tax jurisdiction in the nation.

    2) NJ is next to NYC. Taxes are high in NJ but they are worse in NY. So, a lot of those professionals, especially the ones with children live in NJ.

    3) Once you sweep about a hour driving arc away from NYC, high income earners mostly disappear, and even these are clustered around the rail lines into the city. There is a cluster around Princeton for other reasons of course. Consequently the southern half of the state is poor and depopulated. Areas with good rail service to NY are filled to the brim with high income earners: Monmouth County, the Short Hills area, and the immediate NYC fringe like Hoboken.

  2. Gravatar of Mark A. Sadowski Mark A. Sadowski
    14. February 2011 at 10:59

    Scott wrote:
    “Indeed if tax regimes really didn’t matter, it would be pretty difficult to explain why New Hampshire’s population has grown much faster than any of the other New England states (even given the fact that it is slightly closer to Boston than the other New England states.)”

    I’m not sure population growth is that good an indicator. If I were to look into this I’d check the correlation of the rate of increase in Gross State Product (GSP) per capita with marginal personal income tax rates. Based on the fact that state job growth is correlated with marginal personal income tax rates (so I’ve read) it’s probably also true.

    And New Hampshire may not be as good an anecdote as you think it is. According to this study despite the fact neighboring Vermont has one of the highest marginal personal income tax rates over 1998-2007 it exceeded New Hampshire in both GSP per capita growth and job growth. Vermont evidently has a relatively business friendly environment and a very high quality of life.

  3. Gravatar of Eorr Eorr
    14. February 2011 at 11:31

    California may be much worse tax policy wise but San Francisco is still the best place to live in the US for a lot of reasons.

  4. Gravatar of Bababooey Bababooey
    14. February 2011 at 12:12

    High income taxes go hand in hand with intrusive regulatory environments, entitled public employees, ambitious social justice politicians, and other nasty things that drive businesses away. If you’re immigrating to California to start a business, you will find out soon enough.

  5. Gravatar of dtoh dtoh
    14. February 2011 at 18:07

    One thing people don’t realize is that changes in tax rates have less of an impact on wage earners than they do on capitalists. If you are looking at NYC, you are mostly looking at high wage earners in the finance industry so any impact of taxes is going to be less pronounced than in other regions of the country where high income tends to come from returns on capital.

  6. Gravatar of Scott Sumner Scott Sumner
    14. February 2011 at 19:17

    Jon, Those are good points.

    Mark, I don’t quite agree. I haven’t looked at recent data, but over the past 40 years the population of New Hampshire has grown much faster than any other New England state, including Vermont. It’s not even close. I’d guess Vermont is also well-governed, after all it’s small. And you know what I think of decentralization.

    Regarding income I can say this:

    1. New Hampshire is one of the wealthier states.
    2. New Hampshire has the lowest poverty rate in the US.
    3. New Hampshire has one of the most equal income distributions in the US, maybe the most equal.

    I’d say that’s pretty impressive. New Hampshire is a fairly small town type of state, which usually means lower income. If one argues the high average income is a suburban spillover from Boston, then that begs the question of why income is so equal. Overall, I’d say they do very well.

    Eorr, If it weren’t so cold I’d agree.

    Bababooey, Fortunately I’m moving to LA to retire (in 2017) not start a business.

    dtoh, Yes, the paper I cited said the biggest emigration occurred with those with high capital incomes.

  7. Gravatar of Rien Huizer Rien Huizer
    14. February 2011 at 22:38

    Of course taxes matter, but (1) an attractive location may not lose too much of its attraction by a well designed tax and (b) what do people get in return for those taxes. For instance, if a suburb has particularly good public schools and other services, and high taxes you do not have to send the kids to a private school. Proximity to a very large pool of attractive employment opportunities may have a similar effect. I think such a millionaire tax (a million is not what it used to be) could raise a little revenue without becoming counterproductive in such a place. It might also allow NYC and NY to try to grab a little more..

  8. Gravatar of dtoh dtoh
    15. February 2011 at 01:21

    Does the paper explain why the impact on capital income is more pronounced than on wage income, i.e. the fact that asymmetric returns on investments in new business means that small changes in the tax rate on capital gets hugely amplified in the average expected after-tax return.

  9. Gravatar of Steve Roth Steve Roth
    15. February 2011 at 05:33

    Well here’s the long-run effect. If millionaires vote with their feet, they don’t seem to care about income taxes when they consider what state to live in.

  10. Gravatar of Doc Merlin Doc Merlin
    15. February 2011 at 10:34

    @Steve Roth:

    Well they can afford to pay those taxes… they mostly just hurt the lower middle class, which DO end up having to vote with their feet.

  11. Gravatar of Bababooey Bababooey
    15. February 2011 at 13:56

    San Francisco is still the best place to live in the US for a lot of reasons

    I lived there, loved it, but disagree. Its a great place if you don’t want to grow up and prefer scavenger hunts and dodgeball tourneys to raising children, buying a house and yard, and building a career.

  12. Gravatar of Mo Mo
    16. February 2011 at 07:52

    I wonder if an RD approach with 500K as the arbitrary cutoff might not be a better way to look at this? Wouldn’t answer all the questions – but it would tell us more than diffndiffs.

  13. Gravatar of Scott Sumner Scott Sumner
    16. February 2011 at 18:02

    Rien, NYC is probably near the limit. I’d also like to see them dump income taxes entirely and just tax payrolls.

    dtoh, I don’t recall. I assumed it was because rich people who lived solely off capital income could move to Florida without worrying about losing their job. But that’s just a guess.

    Steve, I’m guessing you are not an economist. Yes, long run effect, but what about the identification problem? If that worked the diff-in-diff study never would have been done.

    Mo, You’ll have to remind me what RD is, I’ve been out of school for 30 years.

  14. Gravatar of dtoh dtoh
    17. February 2011 at 16:45

    No. Because taxes only apply to gain and not losses, the tax impact is hugely impacted when you have assymetric returns. Assume…

    10 Entrepreneurs
    20% Tax Rate
    Each entrepreneur invests $100 ($1000 total investment)

    After some period of time 9 out of 10 entrepreneurs lose their entire investment, but one sees his investment grow from $100 to $1300 for a gain of $1200. The total combined pre-tax rate of return for the 10 entrepreneurs is 30% ($1300/$1000 – 1).

    Since the tax rate is 20%, one might assume that the average after-tax return is 80% of 30% (i.e. 24%) but that is not the case because the tax only applies to positive returns. The losers get nothing back and the one successful entrepreneur pays $240 in tax (20% on the $1200 gain). Therefore as a group the entrepreneurs are left with only a $60 gain on an investment of $1000 or on average a 6% after tax return.

    If the tax is raised to 30% then, the investors who lost money still pay nothing, but the successful investor pays now pays $360. As result after investing $1000, the entrepreneurs are left with $940 which on average works out to a 6% after tax loss.

    With more asymmetric returns even a low tax rate of 15% can cause average expected after-tax returns to become negative. This is why Kennedy/Reagan/Bush tax cuts had such profound effects. Laffer was right even though he did not understand why he was right.

  15. Gravatar of Scott Sumner Scott Sumner
    17. February 2011 at 20:13

    dtoh, I certainly understand your point. I do agree the effects were strong, but I think some supply-siders overstate the effects.

  16. Gravatar of dtoh dtoh
    18. February 2011 at 19:00

    I agree with respect to wage income, but because the supply-siders never understood how the numbers work when you have asymmetric returns on capital, I think they understate the effect on investment, new business formation and overall impact on GDP.

  17. Gravatar of ssumner ssumner
    19. February 2011 at 15:32

    dtoh, I’m not sure they never understood that, I studied it way back in the 1970s.

Leave a Reply