Scott Sumner  

Three big natural experiments

Rogoff on How to Deal with Eme... Origins of the Entitlement Nig...

Natural experiments in economics don't occur very often, at least not big ones of the sort that interest macroeconomists like me. But today we had not one but three big natural experiments. One is already concluded; the other two will play out over the next few months and years.

Experiment #1. Does the powerful real estate lobby have enough political power to prevent Congress from taking away the mortgage interest deduction from the vast majority of taxpayers? Most people previously assumed the answer was yes. But today we found out the answer is no. Under the new tax bill very few taxpayers will deduct mortgage interest.

Experiment #2. Does the mortgage interest deduction play a big role in supporting the price of residential real estate? I suspect the answer is no, but we'll know for sure within a few months.

Experiment #3. Do state income tax rate differentials play a big role in interstate migration? I've argued that state income taxes play a bigger role than many progressives assume, but the effect seems to be declining over time as the younger generation cares more about non-material amenities, rather than material goods like a big house and an expensive car.

I'll watch this one with great interest. Two things to look for:

a. Is there migration from New York and Connecticut to Florida's Gold Coast, or from California to places like Las Vegas, Seattle and Austin?

b. Do politicians in high tax states panic and cut the top rate, in order to hold on to the goose that lays the golden egg?

Both will be fascinating to watch. Kansas was not a test of supply-side economics. This really is a test of supply-side economics.

A few initial reactions:

The good:

1. Smaller marriage penalty.
2. No corporate AMT, and far fewer affected by the personal income tax AMT (which now will be restricted to married couples with incomes over one million (singles over $500,000). That's not very many people. Of course it was stupid not just to eliminate it entirely, and make up the revenue with a tiny boost in the top rate.
3. A limit of $10,000 on SALT deductions. Basically this means that high income people get no deduction for state income taxes, as their property tax will use up most if not all of the $10,000. A huge blow to blue states. (Wait, I'm in California now! Oh well . . . .)
4. Lower marginal tax rates for most people. My family's MTR will fall from 33% to 24%, although of course I'll lose some of that in the limit on SALT deductions.
5. Much higher standard deduction---perhaps the single best change. Far fewer people will itemize.
6. No more health insurance mandate. But I'd label this an incomplete because the GOP will then have to fix Obamacare, and we don't yet know how.
7. Death tax exemption was doubled to $22 million for families. At least my daughter won't have to worry about that one!
8. Corporate rate cut from 35% to 21%
9. Investments can be written off immediately, instead of the complicated and pointless depreciation schedule.
10. Some limits placed on (big) business interest deductibility. I don't know the details on this one yet.

There are probably other good provisions that I don't know about yet---did the go after business lunches? (One of the most outrageous tax breaks)

The bad:

1. Nothing done about health insurance deductibility, the single worst aspect of the tax code. Indeed health expenses are now slightly easier to deduct, but only in 2018. Why? This one is a massive disappointment. Perhaps doctors are more politically powerful than realtors?
2. The 529 education savings account program expanded to cover K-12. This is actually not that bad, but I don't like loopholes.
3. Carried interest loophole maintained? Not certain if I have this one right.
4. Small businesses can deduct interest on loans. Why do we tax equity-financed investments at higher rates than debt-financed investments?
5. Marriage penalty is smaller, but still there. Why?
6. Deductions for quasi-government bonds that finance private activities are maintained.
7. Tax cut will cause the budget deficit to expand sharply---we'll pay a price for this fiscal irresponsibility down the road.
8. Many of the good provisions expire after a few years.


1. Child tax credit increased. I generally don't like loopholes, but this one is pretty simple to handle, and does provide some needed tax relief to lower income families.
2. Deduction for pass-through business. Don't know enough to comment. Could be justified as providing backdoor relief for taxes on investment income, but will rules to prevent wage income diversion add too much to complexity? I don't know.
3. Corporate territorial system. Don't know enough to comment, but we were an outlier with the previous system.

I don't have much to say about the distributional impact, as I don't trust any of the estimates. Distributional effects are an order of magnitude less important than efficiency effects.

Why didn't blue state Republicans object more to the SALT deduction limit? They were bought off with a reduction in the top rate from 43.4% to 40.8% (wrongly reported by the media as 39.6% to 37%) It seems like blue state Republicans care more about affluent Republicans than they do about the states they live in. After all, the cut in the top rate won't in any way discourage people from moving out of New York and California, rather it will simply prevent wealthy taxpayers from being noticeably worse off.

In other words, individual Orange County professionals (my neighbors) will be fine. The state government of California may not be fine.

Comments and Sharing

COMMENTS (23 to date)
E. Harding writes:

"Tax cut will cause the budget deficit to expand sharply---we'll pay a price for this fiscal irresponsibility down the road."
Blame the entire Senate's (except Rand Paul's) willingness to bust spending caps the second it's convenient.

"In other words, individual Orange County professionals (my neighbors) will be fine. The state government of California may not be fine."
Very interesting move. CA/NJ/NY Republicans don't really matter much in the House, though, and there are no genuine blue state Republicans in the Senate.

"Under the new tax bill very few taxpayers will deduct mortgage interest."
I thought $750K was only slightly less than the current $1 million. Can you explain further on this one?

mercer writes:

" Is there migration from New York and Connecticut to Florida's Gold Coast, or from California to places like Las Vegas, Seattle and Austin? "

People have already been migrating because of housing costs. Most people pay a lot more for housing than for state income taxes.

Any migration is a poor test of supply side dogma because of housing costs. The tax plan backers claim corporate tax cuts will increase middle class incomes. Why not use that metric?

Scott Sumner writes:

Harding, The big change is the huge increase in the standard deduction, to $24,000 for married couples. Few will choose to itemize.

Mercer, That will be another interesting test, although it's hard to measure.

On the migration question, I'm thinking especially of high income people. Maybe I should have made that clearer.

Adam J writes:

Scott, isn't the standard deduction change largely muted by the removal of personal exemptions? Getting a $24,000 standard deduction instead of $12,700 seems less good if you can't claim $4,050 in personal exemptions for each of you & your spouse (let alone any kids).

Alan Goldhammer writes:

I'm going to have to wait and see how quickly Turbo Tax gets updated before I can figure out what the personal effects of the tax "reform" bill is. I will take a hit on SALT, don't have any mortgage interest as the house is paid for. The big hits for us start next year when both my wife and I have to take mandatory IRA withdrawals; so much for "lower income" during retirement years.

Scott says he is thinking of high income people in terms of migration. I don't think they are the ones who migrate. It's usually middle income retirees who either want to move to a warmer climate or lower tax state. High income folks living in Manhattan won't leave as they like their lifestyle. Why would anyone want to leave San Diego or San Francisco for Las Vegas?

Several Republican Congressmen from California voted against the House tax bill and probably will vote against the final bill as well (Rohrbacher and Issa for sure as they are likely to face tough reelection races next year). There are some New York and New Jersey Republicans who are also going to be in trouble over this bill. The political folks I know in DC are increasingly of the belief that the House may flip to a narrow Democratic majority next year.

This "reform" is still ridden with too many loopholes and we won't know how the pass through provision works until the accountants and tax lawyers figure out how to game it and it will be gamed just as the effort in Kansas was.

Ken writes:

I’m not sure this new tax bill is about “simplification”. It seems equally complicated but complicated in a different way.

My daughter just finished her third semester of accounting in college. She told me the final exam was mostly about calculating depreciation schedules. But now since capital expenses can be immediately expensed, depreciation schedules are obsolete.

Ken writes:

Prof Sumner made an interesting point about SALT deductions.

Originanally Senators were elected by their home state legislators. But with the 17th amendment, Senators became elected in a popular vote.

I think if it were not for the 17th amendment, this bill would not pass. State legislators would not allow their appointed senator to vote for the elimination of SALT deductions

Dylan writes:


Depreciation will remain relevant for financial accounting and annual reports of public companies.

Alec Fahrin writes:

Not sure why so many people will not use mortgage deduction specifically because of the change from $1,000,000 to $750,000. Maybe you meant the standard deduction getting doubled, but the mortgage deduction itself still applies to 97% of home purchases (instead of the previous 99.5%).

The biggest issue I see with this tax bill is its deficit increase, bad distribution effects, and therefore impending reversal when the Democrats crush the Republican House in 2018. The first order of action by the Democrats will be to demand a new budget that changes this tax bill back to a more progressive distributionary stance, or Trump can risk shutting down the government. That most likely will mean more government deficits (that are always paid for one way or another in the future).

Odd how closely the Obama and Trump first terms are turning out. Difficult initial year, finally pass unpopular legislation, boost the deficit, lose a key senate seat in a freak election, and soon get crushed in the midterms by the opposition. Will Trump get reelected in 2020 somehow? He doesn’t need more than 40% support to win the college.

Thaomas writes:


I agree that mortgage interest should be eliminated because it leads people to substitute consumption of owner occupied housing services for other goods and services. I'm still unclear what consumption distortion is created by the deductability of non-housing property taxes. If it's just a way of generating more revenue from high income people (like retaining the AMT) wouldn't just a higher top rate do the job better?

Scott Sumner writes:

Adam, Yes, but that's a different issue. I'm talking about the effect on the number of people who itemize.

Alan, Yes, California and NYC have amenities, and that allows them to have higher taxes than Kansas. But you need to think at the margin---higher rates will push away at least a few taxpayers.

Alec, You missed the point, it's not about the cut to $750,000, which affects very few people, it's all about the standard deduction. The vast majority of taxpayers will no longer itemize.

And no, you can't win with 40%, at least not unless a third party gets lots of votes.

Thaomas, The deductibility of taxes biases S&L governments toward higher tax rates.

Thomas Sewell writes:

Hard, re: spending caps:
Despite a concerted effort to conflate them, tax cuts aren't the same as government spending.

The Federal government takes in record revenue. Inflation adjusted revenue per person has grown consistently across the decades. The problem is that spending per capita in constant dollars has grown way more.

The budget deficit is a spending issue, not a revenue issue. If you're making more money every year, but spending even more, you don't pass it off as a need to just make even more money. At some point, you have to confront the actual problem, spending.

Brandon Berg writes:

Regarding blue-state Republicans, in general, most people just think about policy in terms of whether their party supports it, or at most the direct effects of the policy. Only a small minority are thinking in terms of indirect effects like migration. But of those who are, maybe they're hoping that limiting the SALT deduction will cause state governments in high-tax states to get spending under control, or at least not increase it so much in the future.

Thaomas writes:


Re SALT deductability

So the "distortion" is that high income people do not resist SALT taxation optimally? I'd prefer to keep tax reform about efficiency and redistribution and leave trying to influence state level politics out of it.

Scott Sumner writes:

Thaomas, This is about efficiency. If states only have to pay 60 cents for each dollar they spend, then they will do infrastructure projects that don't pass a cost/benefit test.

Adam writes:

The effect on high income earners you of limiting the SALT is way overstated. People are getting very little benefit from the SALT because of the pease limitation and the AMT. Take for example a couple with $450k AGI with SALT of $40k. The pease will knock out $4k of the SALT so now the deduction is just $36k. This couple will be subject to AMT because of the SALT and that will add about $7k in tax, equivalent to $20k in deductions at a 35% rate. So now they're effectively getting a $16k deduction, just a bit over the $10k. This couple will likely be way ahead when considering the other changes in the new bill. The difference is that now it's by stealth since most people don't understand how the code works.

Adam writes:

"far fewer affected by the personal income tax AMT (which now will be restricted to married couples with incomes over one million (singles over $500,000)"

I've seen this in a number of article, but it's wrong. Those are the new thresholds for the phase out of the AMT exemption, which is something like $165k. Couples making under$1M can definitely still get hit by AMT under the bill, although it will be much less likely.

Kevin Erdmann writes:

Whatever effect both the MID and SALT have will probably both play out in home values of top tier markets in high cost cities. Might be hard to parse out which effect is responsible for any shifts, even if other effects can be controlled for.

Scott Sumner writes:

Adam, Thanks for the correction in the AMT. Regarding the SALT, I am focusing on a completely different issue from the question of whether people will pay more or less taxes, I'm focusing on the effect on top end MTRs.

Kevin, Don't forget that lots of middle income taxpayers will no longer itemize, so the MID will go away for that group. I'm not saying it will have a big effect on home values, but there is certainly the potential for that group to be impacted.

adam writes:

"Regarding the SALT, I am focusing on a completely different issue from the question of whether people will pay more or less taxes, I'm focusing on the effect on top end MTRs.'

Not sure why this matters. Pease and AMT function as increases to the MTR. The Pease does this in a pretty obvious way: it takes away a dollar of itemized eductions for every 3 cents you earn above the threshold AGI (up to a max of 80% of your itemized deductions). So it effectively adds 3% to the marginal rate. The AMT does the same, although in a much more complicated way.

adam writes:

Argh, I mis-stated that. Should be: "it takes away three cents of itemized deductions for every dollar you earn above the threshold AGI."

Jon writes:

Adam, Scott:

The AMT exemption amount increased to 109K. But the big change is that the exemption phaseout threshold was moved to 400K. That significant moves out the AMT bubble rate which used to hit middle-professional households.

jon writes:
The good:

7. Death tax exemption was doubled to $22 million for families.

I've never understood the obsession with this. Taxes are basically a zero-sum game, so what is the better alternative to an estate tax?
Comments for this entry have been closed
Return to top