Tax Base Broadening ?
Below Dan pulled a post questioning Jon Chait’s claim that US economist’s generally support a broader tax base and lower rates. I am expert neither on taxes nor economists, but I will write at length on the question after the jump.
Before the jump, I must note that Chait made many of the points I try to make below. His main point is that it is easy to propose base broadening in the abstract but hard to eliminate actual deductions. I slipped into criticizing something he wrote in passing in the post Dan pulled over here, and I do it again after the jump.
First I note that many prominent economists support a sharply higher top marginal income tax rate. The figure 70% is not rare. I am not prominent, but that happens to be the top marginal income tax rate which I support (from nostalgia — that was the rate during my childhood in the 60s a period of notoriously slow economic growth).
Second, I think the consensus is a villager VSP consensus about what economists say (or should say or what economic theory implies or something). Similarly the VSPs are sure that economics tells us that social security pensions should be cut. Oddly many actual economists disagree.
Third, I support gradually eliminating the mortgage interest deduction, so, I personally also support base broadening (in addition to raising the top rate). Here I will generally argue against base broadening, but I stress that I am making a case against it and not totally opposed.
Now for some discussion of actual policy.
I think that it is easy to oppose tax expenditures (deductions, credits, exclusions etc) in the abstract for the same reason it is easy to support lower government spending in the abstract. All are costly, but the justification for each is specific, so when one just thinks of the whole mess, one thinks of the costs and not the benefits.
I don’t think that Chait thinks that most economists want to eliminate the earned income tax credit. I’m sure he doesn’t. That’s a tax credit, but it isn’t what he means at all. Same for the child tax credit.
My guess is that he wants to tax employer provided health insurance as income, but only a part of a transition to single payer. Certainly, he isn’t for repealing the ACA and eliminating the deduction. The logic of a broad tax base is to minimize distortions due to decisions made to avoid taxes. But the huge distortion here is the problem of free riding by the uninsured. Employers who pay for health insurance pay for the care of employees of employers who don’t. Chait supports a mandate — that is an additional tax incentive to get health insurance.
Some economists might want to eliminate tax incentives to save such as IRAs an Keoghs (does anyone remember what Keoghs are ? My dad has one). But wait the problem with running deficits is that is that the illusion of wealth causes higher consumption which crowds out investment (only if the economy is not in a liquidity trap).
Loopholes in the corporate tax code are proportionally much more important than loopholes in the personal income tax code. But many make a whole lot of sense. Surely we prefer corporations to reinvest profits rather than paying dividends (again the problem with just running deficits is crowding out investment). We should especially like investment in R&D since the investing corporation does not capture all of the benefits.
Hmm I think I have covered a significant fraction of total tax expenditures. So what is the cause of the near consensus for a broader tax base and lower rates(among VSPs at least). Clearly a lot of it is that they don’t find the alternative attractive — that is if one is convinced that high tax rates have high dead weight losses, then one is morewilling to eliminate tax expenditures. The alleged costs of high rates include avoiding genuine income to avoid paying taxes — reducing labor supply, reducing human capital investment and reducing saving. The evidence for a large effect on labor supply is miniscule. The human capital argument is that people study less if the increased salary mostly goes to taxes. Again the evidence is very weak. The remaining issue relates specifically to taxation of capital income and capital gains. I think the special low rates are considered loopholes by the critics of loopholes. I know some economists claim to have calculated significant dead weight losses due to genuine avoidance of genuine income. I know how much they had to torture the data too. But I don’t think that anyone is really convinced.
Rather, the problem with high rates is that they encourage tax avoidance for given gross income by making deductions and shelters and such like more valuable. So we are back to deciding if such avoidance (which is often explicitly encouraged) is desirable. It is generally the product of a political choice to encourage something. If this was a mistake, it is a mistake for low tax rates too. If it is a good policy, high rates make it more effective.
In any case, a higher tax rate does not imply that deductions must be more valuable. It is possible to cap deductions (as Obama has repeatedly proposed) so, for income over $250,000 the deduction of x is replaced by a credit of 0.35x. The incentive to find it is the same as now. The new top tax rate can be anything. There is no reason at all taxes have to be calculated by adjusting income then applying the tax function. The almost universal conflation of the marginal tax rate and the marginal benefit to the taxpayer of a tax expenditure is not an error — it is a trick used by those who want low tax rates to suggest that lowering rates is the only way to reduce incentives to find deductions and shelters and such.
OK now for my one field of expertise — lazyness. I am not the world’s top expert, but people even lazier than me are too lazy to type. I suspect that pundits (and some economists) consider the current tax code inefficient, because they personally would have to work rather hard to commment on it intelligently (my solution as seen above is to comment on it based on guesswork). The efficiency is dollars raised per page of tax code. The wasted effort due to complexity is their effort. I think this is a factor.
I think another factor is the desire to find common ground. A compromise based on lower rates and fewer loopholes appeals to lovers of bipartisanship. They have nostalgia for the distant memory of Bill Bradley and Jack Kemp working together. In particular, the argument against high rates based on incentives for tax avoidance leads invevitably to the proposal that the loopholes be eliminated. This is a way to increase revenues which supply siders must embrace if they don’t want to be revealed to be total hypocrites. Decades of experience have not taught many VSPs that supply siders don’t mind broadcasting their total hypocrisy.
Finally, I think advocacy of a simpler tax code appeals to people because it seems more horizontally equitable. The logic is not utilitarian. The idea is that it is fair for people with the same gross income to pay the same amount in taxes. I don’t have any patience for this argument (I don’t care about horizontal equity — I care about equality). It is based on the asusmption that, because it seemed once long ago that income was a good measure of ability to pay taxes, it is the only fair measure. If you think the argument makes sense, think of the marriage penalty. Should the marginal tax rate depend on individual income or family income ? They can’t both have some fundamental link with fairness.
I think the extremely strong appeal is that everyone thinks of the clever sneaks who are paying less than we are (not of those even less sophisticated than ourselves who are paying more). The thought that someone somewhere has more money to spare, yet pays less in taxes, is infuriating. This is not a rational basis for public policy.
But by all means get rid of the mortage interest deduction, slowly, with plenty of warning, when the economy is back to normal.
“Surely we prefer corporations to reinvest profits rather than paying dividends”
That’s what I used to think. But then I found out that a pretty good investment strategy is to buy companies with moderately high dividends. Maybe people can handle money better than corporations. 🙂
I agree with most of what this post says but why not simply change the cap on the home mortgage deduction to something far lower than today’s cap? The current cap: interest on one million dollars in loans, more than four times the median price of a new home. The high cap is what makes this deduction so valuable to people with high incomes, rather than the middle class. It could be halved and then halved again, for example. Most people believe that the rich get all the deductions and, more than that, that businesses deduct virtually all their so-called expenses, and they won’t appreciate being the target for cuts in “tax expenditures” as defined by some politician (or economist?).
Min yes high dividend yield is correlated with high returns. But by “we” I mean we citizen tax code writers. As such we want to encourage investment (to make up for capital income taxes if nothing else). Except now when the economy is in a liqudity trap, I don’t want you to spend your dividends, so I’d rather corporations didn’t give you a chance to do so. This just enough for me to support the investment tax credit.
PJR You are absolutely righ that it is politically impossible to eliminate the mortgage interest deduction. I think you are also right that it can be capped. So I totally agree with your proposal.
actually i thought robert was on to something when he said
the justification for each is specific, so when one just thinks of the whole mess, one thinks of the costs and not the benefits.
this is a pretty good description of thinking in general. people think of what they think of, and forget about what they don’t think of.
i think robert provides pretty good examples of this in the rest of his post, but I think he realizes it.
me, i tend to favor the simplest tax code possible… moderately progressive… that is, according to ability to pay. most of the arguments for your favorite deduction would, i think, melt away when you realize, for example, that you don’t get the benefit of the mortgage deduction. the bank does.
don’t take that example too seriously… i dont think i am capable of the kind of detailed analysis that would prove it…. but i have a feeling that whatever the taxes are, the economy adjusts to it.
for example, if i never heard of taxes i would be happy to make 80k on some deal, but since i know i will pay taxes, i will ask for 100k, and the other party will pay it, because he knows about taxes too.
re health insurance deduction:
i don’t think “the problem” is free riding. i think it’s pay or die pricing. something that might not be as lucrative for the provider were it not for insurance. and of course insurance would not be so profitable if medical costs were not so high.
if we are worried about free riding we would have a dedicated, “flat”, tax up to a reasonable cap, and stop giving the providers and insurance companies a free ride. the government could either pay the providers, or contract the daily management to the insurance companies, or establish government run health clinics in competition with the private boys… plenty of better ways than giving a tax cut to the employer or hitting the individual with a mandate to go get screwed by the established crooks.
there is no reason to favor “reinvestment” over dividends. that is a decision safely left up to the corp and its potential competitors. “reinvestment” just sounds good because it promotes (it says here) “growth,” and we all know that growth is what we need more of, at all costs.
I am with Robert re generally favoring reinvestment over dividends. It’s not a bad idea to see more expansion/hiring, for example, and less in checks to wealthy investors. The latter would own a more valuable company, so to them it’s mostly deferring a little gratification.
Robert Waldmann: “Min yes high dividend yield is correlated with high returns. But by “we” I mean we citizen tax code writers. As such we want to encourage investment (to make up for capital income taxes if nothing else).”
Yes, we are on the same page about intentions. But if corporations were able to make effective investments, given the existing tax breaks for doing so, wouldn’t that be reflected in the return on the shareholders’ investment in the corporation? IIUC, most economists believe that the private sector is good at investment. If so, then those tax breaks should pay off, no?
I may not have expressed myself clearly. Dividends, after all, stay in the private sector, too. But if I am an investor using a high yield strategy, I will be putting a lot of those dividends to work. If corporations could put that money to work better, and get a tax break for doing so, why would a high-yield strategy work? Perhaps it is a question of diminishing returns. The extra investment by the corporation that the tax break produces may not be worth as much as other investment opportunities that are available.
You might invest the dividends in other firms, but you might spend them. I am assuming we want to encourage saving. If firms reinvest profits, investors as a group are forced to save. I think this is socially desirable for many reasons one being to undo the effect of capital income taxation.
On dividend yield and return on shares you seem to be switching the efficient markets hypothesis on and off – you don’t believe it because you think you have found a better strategy than buy and hold, but you do believe it because you think that shares will increase in value by the true value of the new capital produced by investment.
One way a firm can have a low dividend yield is if the price of iy’s shares are high for no good reason. The number which I know to be correlated with high returns is the ratio of dividends to share prices. I don’t know about the correlation of the ratio of dividends to profits and subsequent returns,
Why “moderately progressive” why not extremely progressive ? I favor a highly progressive tax code. The logic is the same as the case for moderately progressive. To stop at moderately one needs an explanation of bad effects of high marginal tax rates. The only bad effects for which there is evidence is increased tax avoidance via loopholes etc. If the tax code were simple ( or deductions corrections credits etc were capped) I would see no reason the tax code shouldn’t be extremely progressive.
The logic is pure add up the utils utilitarian — taking a dollar from a richer person hurts less as he has a lower marginal utilty of consumption. This suggests high optimal top marginal rates over 70%.
This is consistent with my post. Better than the current US system is easily possible. But repealing Obamacare and taxing employer contribution to health insurance would be even worse than just repealing Obamacare (the debate was originally about Paul Ryan’s proposal).
There are many good reasons to promote high investment. One is the very strong evidence that the same worker gets a higher wage if he works with more capital. In simple standard theory this shouldn’t happen, but the evidence is very strong. This means that the firm and its shareholders do not capture all of the benefits from investment, so the share value maximizing level is lower than the socially optimal level.
Also taxes on capital income favor consumption over invesmtent.
Also Romer 86 spillovers for which there is some evidence.
You are assuming that the market works optimally. This requires strong assumptions which are inconsistent with the data.
Largely agree with what you wrote. Added points:
1. My employer share of health care was $10K, mine was $5K. $15K is a pretty large tax on my income. I’m not sure what an additional real tax of 5% for universal health care would be objectionable. What I mostly hear is “I don’t want some bureaucrat between me and my doctor”. My response “so you prefer an insurance agent who’s primary task is profit for the comapny”.
2. Recent report by Jeffrey Thompson at UMass Ahmerst – higher taxes don’t cause rich people to move. With all the other previous studies like Diamond-Saez. I like 70% too.
3. Lets be bold if we want to give people something for having kids, lets give it to them – all of them. Not as a tax credit, but as a check in the mail. Some for house ownership or any other things we deem are important. (personally I think we should not incentivize house ownership).
4. And even bolder, lets tax consumption and let people keep their first $50K without taxes, then start adding surtaxes as earnings go up. And lets not forget to tax wealth transfer. Concentration of wealth in certain families is not generally a good thing for broader society.
Yeah, I might be radical, but if we’re going to negotiate, might as well start equally far away from the middle as the other side 🙂
I agree with you entirely. Especially that refundable credits are better than deductions and a steeply progressive tax on consumption (with rates way over 100% so if you spend 1 million you send the IRS 10 million is a good approach). Finally, I agree that idf we shift to progressive consumption taxationm we need a high estate tax.
Overall, good essay!
Soebody told me recently that “The only thing that I believe from a management team is the quarterly dividend. Everything else is BS.” Why not let the OWNERS decide what to do with profits?
My thought exactly: limit the amount of debt on which the interest can be deducted, or simply limit the amount of interest that can be deducted. The first avoids the problem that people who had lower interest rates would be able to deduct a larger proportion of their interest — but is sorta complicated to calcluate, though banks could simply print the relevant amount on the appropriate form they send out anyway — the math is simple enough for computers: here is total debt this year; here is the interest rate; here’s the deductible amount. The second approach would be very easy to calculate.
In a (negative) Pigovian and public policy sense, even better would be a sliding deduction based on income, like the amount of Roth contribution one can make. I’m willing to allow a modest subsidy for people with lower incomes, but why offer any interest subsidy to the wealthy? The bubblicious effects are by now pretty clear.
So: up to $4000 of mortgage interest debt is deductible on an income under $100,000; above that, the maximum amount deductible decreases to $0 at $200,000 of income, all adjustable for inflation yearly by the IRS.
PQQuincy the cap and the complicated calculation already exist. Few know about it because few get close to the $1 million cap. If you reduce the cap substantially, more people would need to perform the complicated calculation, but I doubt that many of these homeowners are doing their taxes without software (or accountants).
Robert Waldmann: “On dividend yield and return on shares you seem to be switching the efficient markets hypothesis on and off – you don’t believe it because you think you have found a better strategy than buy and hold, but you do believe it because you think that shares will increase in value by the true value of the new capital produced by investment.”
I think that the weak EHS is true in a long enough run if enough things remain enough the same. 😉 It is not difficult to find exceptions in the short run, but not easy to find profitable exceptions. I first heard about the high yield strategy in the mid-1980s. I was surprised to hear recently that it is still a good strategy. I would have expected it to have self-destructed by now, by the weak EHS. If it has not, then I suspect that it is because it is unpopular. It certainly is not a strategy with a hot sell, unlike the promise of sky high returns and the peddling of risk.
But if it is a good strategy, then how good is reinvestment by firms? (I did not compare a high yield strategy with buy and hold, BTW.)
“The thought that someone somewhere has more money to spare, yet pays less in taxes, is infuriating. This is not a rational basis for public policy.”
actually it’s not a good basis for personal sanity either. of course “sanity” is an old fashioned word, but it replaced the even more old fashioned one you would find even more offensive.
we can probably all afford to pay a reasonable share…based on ability to pay… for the services we, as a nation, think we need from the government. we don’t need to be infuriated because we think someone else is getting the biggest piece of the burfday cake.
of course we could try to design a system that was indeed more “fair.” probably we could even agree about that if we could get over … well, it seems to go both ways.
As you note you were “surprised”. You shouldn’t have been. Belief in the weak EMH is like belief in the flat earth hypothesis. The same patters have been in the data for decades.
In any case, you definitely have contradictory beliefs. You are using the EMH and claim (correctly) to have found a profitable anomaly. One or the other.
“long enough run” does not go with h for hypothesis. A hypothesis should be testable,which means testable with a finite amount of data. Also, you are making claims about what is socially desirable using the EMH. If it were to become valid in a billion years, that would not mean that optimal policy is based on assuming it is valid now. Long enough run is not only inconsistent with any effort to be scientific, it is inconsistent with any desire to say anything potentially useful to anyone.
You will note that I have very very strong feelings about “long enough run.” My motto is “asymptotically we will all be dead.” I am absolutely unable to maintain normal standards of politeness with people who write “long enough run” and I have no intention of trying. I think typing that phrase is a grave sin against science and truth.